INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

repeated references to the name "Harris."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

rious

-0.77

unal

-0.74

yip

-0.73

nces

-0.72

erous

-0.68

culosis

-0.67

ition

-0.66

gencies

-0.65

 Yugoslav

-0.65

uesday

-0.64

POSITIVE LOGITS

burg

1.09

 Harris

0.95

Harris

0.95

mann

0.88

acre

0.81

laughter

0.80

iland

0.77

ippi

0.77

VILLE

0.77

burgh

0.76

Activations Density 0.005%

proper nouns

repeated references to the name "Harris."

No Comments

No Known Activations