INDEX

Explanations

phrases refering to general human populations or individuals

oai_token-act-pair · gpt-3.5-turbo

phrases related to general human experiences and perspectives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Samson

-0.68

 Lithuania

-0.64

 Spiegel

-0.64

srf

-0.60

 Polish

-0.59

hua

-0.58

 storm

-0.58

 Orchestra

-0.58

 Baltic

-0.58

 teasp

-0.57

POSITIVE LOGITS

 imaginable

1.05

 EVER

0.83

 besides

0.76

 revolves

0.75

WER

0.75

oppers

0.74

 alike

0.72

ever

0.72

 ever

0.70

WN

0.70

Activations Density 0.253%

phrases refering to general human populations or individuals

phrases related to general human experiences and perspectives

No Comments

No Known Activations