INDEX

Explanations

proper nouns and names, specifically focusing on "hart" and "Friedman"

oai_token-act-pair · gpt-3.5-turbo

repeated mentions and references to specific individuals with the last name "Hart" and "Friedman."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

Loading

-0.78

terday

-0.76

ision

-0.76

ple

-0.74

ively

-0.73

staking

-0.73

itals

-0.70

ISION

-0.70

Mill

-0.69

lect

-0.69

POSITIVE LOGITS

 Ducks

0.76

 Friedman

0.76

ãĤ¤ãĥĪ

0.72

awi

0.69

 Vive

0.67

Jr

0.66

zman

0.65

 Airl

0.65

)=(

0.64

 Palmer

0.64

Activations Density 0.035%

proper nouns and names, specifically focusing on "hart" and "Friedman"

repeated mentions and references to specific individuals with the last name "Hart" and "Friedman."

No Comments

No Known Activations