INDEX

Explanations

the word 'that' with high relevance

oai_token-act-pair · gpt-3.5-turbo

the word "that" 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 @ 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

Returns

-0.74

parts

-0.68

oses

-0.68

EngineDebug

-0.63

thumbnails

-0.60

laughs

-0.60

Is

-0.59

iHUD

-0.59

roots

-0.58

Beg

-0.57

POSITIVE LOGITS

 aren

1.31

 defy

1.22

are

1.20

 involve

1.17

 resemble

1.14

 rely

1.12

 seem

1.10

 originate

1.09

 emphasize

1.08

 weren

1.06

Activations Density 0.167%

the word 'that' with high relevance

the word "that" in various contexts

No Comments

No Known Activations