INDEX

Explanations

words related to commuting or official military positions

oai_token-act-pair · gpt-3.5-turbo

terms related to community engagement and communication

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

other

-0.66

 Sloven

-0.66

 nerv

-0.64

unders

-0.63

Euro

-0.61

 Feldman

-0.61

 Finnish

-0.61

under

-0.60

 Painter

-0.60

 Eisen

-0.60

POSITIVE LOGITS

rals

0.80

ruary

0.79

rencies

0.78

iton

0.76

issance

0.73

urations

0.73

uted

0.73

rocal

0.70

berus

0.69

urate

0.69

Activations Density 0.030%

words related to commuting or official military positions

terms related to community engagement and communication

No Comments

No Known Activations