INDEX

Explanations

phrases related to actions and consequences

oai_token-act-pair · gpt-3.5-turbo

references to games, health, and actions involving loss or gain

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

than

-0.79

\",

-0.61

inventoryQuantity

-0.60

albeit

-0.58

Too

-0.57

reenshot

-0.57

 è£ıè

-0.56

)\

-0.55

reenshots

-0.53

@#&

-0.53

POSITIVE LOGITS

 forever

1.00

 thereafter

0.96

 whichever

0.94

 indefinitely

0.89

 accordingly

0.88

.[

0.81

 until

0.81

 thereto

0.79

 respectively

0.77

 hereafter

0.76

Activations Density 0.864%

phrases related to actions and consequences

references to games, health, and actions involving loss or gain

No Comments

No Known Activations