INDEX

Explanations

mentions of specific organizations or locations mentioned in news articles

oai_token-act-pair · gpt-3.5-turbo

references to the New York Times

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 advis

-0.74

wagen

-0.73

 aquarium

-0.70

 debtor

-0.70

 purse

-0.69

 induct

-0.68

 stewards

-0.68

 enthusi

-0.68

 guiActiveUn

-0.67

imus

-0.66

POSITIVE LOGITS

Va

0.83

0.80

0.79

Org

0.77

org

0.77

jpg

0.75

CO

0.75

0.74

0.73

Activations Density 0.097%

mentions of specific organizations or locations mentioned in news articles

references to the New York Times

No Comments

No Known Activations