INDEX

Explanations

instances where the document mentions the act of thinking or reflecting

oai_token-act-pair · gpt-3.5-turbo

discussions centered around the concept of thinking

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

 shale

-0.66

catentry

-0.65

bri

-0.62

waters

-0.58

 contamin

-0.58

CBC

-0.57

key

-0.57

 abolished

-0.57

ague

-0.55

videos

-0.55

POSITIVE LOGITS

orial

0.82

 Turing

0.77

-|

0.77

pad

0.76

lift

0.72

ortment

0.70

onymous

0.70

xiety

0.69

 aloud

0.68

iste

0.68

Activations Density 0.046%

instances where the document mentions the act of thinking or reflecting

discussions centered around the concept of thinking

No Comments

No Known Activations