INDEX

Explanations

mentions of the mainstream media

oai_token-act-pair · gpt-3.5-turbo

repeated references to mainstream media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

arcity

-0.83

atoon

-0.82

cific

-0.81

 Tome

-0.73

otos

-0.70

tein

-0.69

thur

-0.68

uana

-0.68

alid

-0.68

haps

-0.67

POSITIVE LOGITS

 media

0.88

 outlets

0.85

ization

0.84

 mainstream

0.79

iary

0.78

arily

0.78

ership

0.76

ing

0.74

 wisdom

0.73

isation

0.72

Activations Density 0.039%

mentions of the mainstream media

repeated references to mainstream media

No Comments

No Known Activations