INDEX

Explanations

mentions of outlets, often in the context of media or information dissemination

oai_token-act-pair · gpt-3.5-turbo

references to media outlets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

heed

-0.72

 pudding

-0.66

jri

-0.65

ayne

-0.62

eric

-0.62

grave

-0.62

ening

-0.61

ee

-0.60

sis

-0.59

 Defin

-0.59

POSITIVE LOGITS

 outlets

0.91

 outlet

0.90

="#

0.80

swick

0.74

Wem

0.73

obo

0.70

 CCTV

0.68

eval

0.68

 SPORTS

0.66

oway

0.64

Activations Density 0.053%

mentions of outlets, often in the context of media or information dissemination

references to media outlets

No Comments

No Known Activations