INDEX

Explanations

locations or positions

oai_token-act-pair · gpt-3.5-turbo

locations or references to unspecified places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

iger

-0.76

DOM

-0.73

advertisement

-0.71

runners

-0.69

ulators

-0.68

iff

-0.67

report

-0.67

DOS

-0.67

iasis

-0.66

abilities

-0.66

POSITIVE LOGITS

 else

1.48

Else

1.35

 Else

1.08

abouts

1.03

else

0.92

ãĤ´ãĥ³

0.79

 Somewhere

0.77

sterdam

0.77

ĪĴ

0.75

 unspecified

0.75

Activations Density 0.023%

locations or positions

locations or references to unspecified places

No Comments

No Known Activations