INDEX

Explanations

keywords in a list format, specifically focusing on the phrases "takeaways," "bits," "facts," "myths," "recap," "highlights," "notes," and similar terms

oai_token-act-pair · gpt-3.5-turbo

key takeaways and insights from a discussion or report

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

endars

-0.78

gur

-0.75

ebook

-0.73

psc

-0.73

azo

-0.72

abella

-0.70

raph

-0.70

raid

-0.69

esi

-0.69

urated

-0.66

POSITIVE LOGITS

 pertaining

0.93

 glean

0.86

 regarding

0.86

:-

0.85

 relating

0.81

":[{"

0.75

 related

0.75

 concerning

0.72

 pertinent

0.68

viz

0.68

Activations Density 0.229%

keywords in a list format, specifically focusing on the phrases "takeaways," "bits," "facts," "myths," "recap," "highlights," "notes," and similar terms

key takeaways and insights from a discussion or report

No Comments

No Known Activations