INDEX

Explanations

references to individuals named "Lew" or "Lewinsky"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Lew" and related references

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ACTED

-0.83

CHR

-0.77

senal

-0.69

 bulls

-0.69

 à¨

-0.69

expr

-0.66

Js

-0.63

 Future

-0.62

GROUP

-0.62

PDATE

-0.61

POSITIVE LOGITS

andowski

1.42

insky

1.19

inski

0.96

enstein

0.93

die

0.86

ield

0.84

ington

0.84

avering

0.83

iky

0.83

itsch

0.82

Activations Density 0.039%

references to individuals named "Lew" or "Lewinsky"

mentions of the name "Lew" and related references

No Comments

No Known Activations