INDEX

Explanations

questions or statements starting with the word "Where"

oai_token-act-pair · gpt-3.5-turbo

the word "Where"

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

ÂłÂłÂłÂłÂłÂłÂłÂł

-0.70

)].

-0.70

 roller

-0.61

 franc

-0.57

 secretion

-0.56

 gripping

-0.55

 absorbing

-0.54

 swipe

-0.54

 palp

-0.53

<+

-0.53

POSITIVE LOGITS

fore

1.49

upon

1.45

abouts

1.43

ver

1.34

soever

1.07

velt

0.78

Dat

0.76

with

0.76

 else

0.76

ipl

0.76

Activations Density 0.047%

questions or statements starting with the word "Where"

the word "Where"

No Comments

No Known Activations