INDEX

Explanations

words related to physical actions or events, often involving force or impact

oai_token-act-pair · gpt-3.5-turbo

references to emotions and conflicts in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ITNESS

-1.11

itness

-0.65

vant

-0.56

æĢ

-0.55

 millenn

-0.55

 Surveillance

-0.54

asia

-0.54

å¤©

-0.54

itable

-0.53

 Might

-0.53

POSITIVE LOGITS

 into

1.42

 away

1.36

 onto

1.32

 down

1.28

out

1.27

 apart

1.23

 overboard

1.21

off

1.19

 aside

1.17

 goodbye

1.13

Activations Density 0.750%

words related to physical actions or events, often involving force or impact

references to emotions and conflicts in various contexts

No Comments

No Known Activations