INDEX

Explanations

words related to interference or congestion

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "jam" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Nations

-0.72

 Affairs

-0.69

 reconc

-0.68

 Correspond

-0.67

 Ones

-0.65

 Owners

-0.64

 Advantage

-0.64

ģ«

-0.64

 Incarnation

-0.63

 Highlander

-0.63

POSITIVE LOGITS

ming

1.48

mers

1.08

bling

1.01

mer

0.98

jam

0.95

ieu

0.92

ilton

0.91

bly

0.86

jam

0.86

weed

0.85

Activations Density 0.006%

words related to interference or congestion

occurrences of the word "jam" and its variations

No Comments

No Known Activations