INDEX

Explanations

phrases related to mutual interaction or agreement

oai_token-act-pair · gpt-3.5-turbo

concepts related to mutual relationships and cooperation

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

teenth

-0.76

dq

-0.76

nor

-0.73

nai

-0.71

HOU

-0.66

mble

-0.66

mb

-0.66

————

-0.65

lay

-0.64

 Mandatory

-0.64

POSITIVE LOGITS

 admiration

0.95

 acquaintance

0.90

 masturbation

0.89

aid

0.89

itÃ©

0.87

 respect

0.85

 understanding

0.82

 benefit

0.81

 agreement

0.80

 distrust

0.78

Activations Density 0.060%

phrases related to mutual interaction or agreement

concepts related to mutual relationships and cooperation

No Comments

No Known Activations