INDEX

Explanations

words related to high points or peaks

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "peak" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

icate

-0.81

oured

-0.75

onsense

-0.73

 subtitles

-0.72

signed

-0.68

ãĥĻ

-0.68

avering

-0.66

 Cosponsors

-0.66

ilage

-0.65

aves

-0.63

POSITIVE LOGITS

lite

0.74

 hour

0.72

 occupancy

0.71

imum

0.69

 periods

0.69

Activity

0.68

 Performance

0.67

antha

0.66

yip

0.65

iphany

0.65

Activations Density 0.078%

words related to high points or peaks

occurrences of the word "peak" in various contexts

No Comments

No Known Activations