INDEX

Explanations

phrases indicating additional information or quotes added to a statement

oai_token-act-pair · gpt-3.5-turbo

instances of the verb "added" and related forms, indicating the inclusion of information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 mathemat

-0.61

prototype

-0.57

misc

-0.56

inances

-0.56

folk

-0.55

spread

-0.55

pod

-0.54

gone

-0.54

 Compact

-0.53

cies

-0.52

POSITIVE LOGITS

 omin

0.92

 sarcast

0.88

ictions

0.85

 insult

0.81

itional

0.79

:"

0.78

0.77

endum

0.76

 bluntly

0.76

"-

0.73

Activations Density 0.039%

phrases indicating additional information or quotes added to a statement

instances of the verb "added" and related forms, indicating the inclusion of information

No Comments

No Known Activations