INDEX

Explanations

phrases related to specific actions or events

oai_token-act-pair · gpt-3.5-turbo

the word "but"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Aval

-0.78

 ingred

-0.69

visor

-0.67

 masses

-0.65

 transcription

-0.64

 scen

-0.64

 harmless

-0.63

 metic

-0.63

 managerial

-0.63

izational

-0.63

POSITIVE LOGITS

ERC

0.81

 Trident

0.77

CBC

0.76

 Sapphire

0.71

ARI

0.71

aren

0.70

Hamilton

0.69

AG

0.69

enna

0.69

ELL

0.67

Activations Density 0.000%

phrases related to specific actions or events

the word "but"

No Comments

No Known Activations