INDEX

Explanations

words related to sudden or intense actions

oai_token-act-pair · gpt-3.5-turbo

words related to physical interactions or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

resso

-0.68

lander

-0.66

elt

-0.65

omet

-0.65

lar

-0.62

alt

-0.60

astical

-0.58

opens

-0.58

erest

-0.58

ruby

-0.58

POSITIVE LOGITS

kefeller

0.77

ROR

0.74

SPONSORED

0.70

doors

0.69

@@@@@@@@

0.65

VICE

0.64

 Dragonbound

0.63

 Warfare

0.62

liness

0.61

xual

0.61

Activations Density 0.103%

words related to sudden or intense actions

words related to physical interactions or actions

No Comments

No Known Activations