INDEX

Explanations

specific mentions of television channels

oai_token-act-pair · gpt-3.5-turbo

mentions of television channels

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

berman

-0.69

 Jarrett

-0.65

ggies

-0.64

enhagen

-0.62

stad

-0.61

 Summers

-0.60

dden

-0.60

hya

-0.60

 Arkham

-0.59

 Raiders

-0.59

POSITIVE LOGITS

lain

0.84

 Divinity

0.83

agall

0.82

 surfing

0.77

ed

0.77

izen

0.76

 channel

0.76

nel

0.75

ages

0.74

 Tunnel

0.72

Activations Density 0.047%

specific mentions of television channels

mentions of television channels

No Comments

No Known Activations