INDEX

Explanations

the occurrence of specific numerical quantities of people performing various actions or being involved in events

oai_token-act-pair · gpt-3.5-turbo

quantifiable data about populations or significant groups of people

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

inventoryQuantity

-0.82

Nid

-0.66

iste

-0.65

SourceFile

-0.65

 superiors

-0.64

 selection

-0.63

leader

-0.62

Leader

-0.61

 Selection

-0.61

feeding

-0.59

POSITIVE LOGITS

 crammed

0.98

 annually

0.94

 perished

0.85

 worldwide

0.83

 collectively

0.81

 illegally

0.79

 converge

0.77

 stockp

0.73

 stranded

0.72

 conver

0.70

Activations Density 0.375%

the occurrence of specific numerical quantities of people performing various actions or being involved in events

quantifiable data about populations or significant groups of people

No Comments

No Known Activations