INDEX

Explanations

English words

oai_token-act-pair · gpt-3.5-turbo

instances of the word "En"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

PAC

-0.75

 stretch

-0.74

 remission

-0.69

AMA

-0.69

 pound

-0.68

 bonding

-0.68

 circulation

-0.68

 sport

-0.66

 sports

-0.65

 summer

-0.64

POSITIVE LOGITS

En

3.39

Enc

1.58

Ut

1.28

EN

1.27

Armor

1.24

Il

1.20

Requ

1.19

Ob

1.18

Inst

1.17

Eng

1.17

Activations Density 0.010%

English words

instances of the word "En"

No Comments

No Known Activations