INDEX

Explanations

proper nouns with an organizational structure or abbreviation

oai_token-act-pair · gpt-3.5-turbo

parenthetical phrases or references

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

icy

-0.73

quished

-0.72

 baffled

-0.72

wre

-0.71

 haunted

-0.69

 haun

-0.69

lyr

-0.69

 resur

-0.68

lyn

-0.68

 pale

-0.68

POSITIVE LOGITS

formerly

1.32

including

1.25

sic

1.24

aka

1.16

excluding

1.14

see

1.13

which

1.12

...)

1.09

?),

1.07

emphasis

1.07

Activations Density 0.154%

proper nouns with an organizational structure or abbreviation

parenthetical phrases or references

No Comments

No Known Activations