INDEX

Explanations

mentions of the word "Bro" in the text

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "Bro" or related references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

aneous

-0.71

 Nadu

-0.65

urers

-0.65

lessness

-0.61

 repetition

-0.61

 innocence

-0.61

 liberty

-0.60

aries

-0.60

URA

-0.60

orship

-0.59

POSITIVE LOGITS

ccoli

1.25

oks

1.16

thel

1.12

dy

1.12

gue

1.02

chet

0.99

keye

0.98

oke

0.97

thren

0.95

swer

0.95

Activations Density 0.027%

mentions of the word "Bro" in the text

occurrences of the term "Bro" or related references

No Comments

No Known Activations