INDEX

Explanations

words containing the sequence "dist" or "dist" followed by a number

oai_token-act-pair · gpt-3.5-turbo

words related to distress or negative emotional states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

tes

-0.87

swick

-0.82

glers

-0.82

theless

-0.76

ton

-0.72

ggle

-0.70

phone

-0.70

LOAD

-0.69

FORE

-0.68

FIELD

-0.65

POSITIVE LOGITS

anced

1.18

ribut

1.17

illery

1.01

ancing

1.00

 dist

0.95

aste

0.95

ributes

0.94

ances

0.88

enfranch

0.88

antly

0.86

Activations Density 0.007%

words containing the sequence "dist" or "dist" followed by a number

words related to distress or negative emotional states

No Comments

No Known Activations