INDEX

Explanations

action verbs indicating personal experiences or actions

oai_token-act-pair · gpt-3.5-turbo

past tense verbs indicating actions or experiences

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

 dissip

-0.63

 imped

-0.63

Appearances

-0.61

 endangered

-0.60

}}

-0.58

 obsc

-0.57

 liabilities

-0.57

 obstruct

-0.57

})

-0.56

Ø©

-0.56

POSITIVE LOGITS

 myself

1.31

my

0.87

nesday

0.78

escription

0.74

eah

0.73

ragon

0.70

ucc

0.70

yss

0.69

ocument

0.69

displayText

0.66

Activations Density 0.324%

action verbs indicating personal experiences or actions

past tense verbs indicating actions or experiences

No Comments

No Known Activations