INDEX

Explanations

text related to embedding codes within a document

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "embed" and its variations in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

externalActionCode

-0.87

 Sabha

-0.86

veyard

-0.84

ĸļ

-0.77

iard

-0.72

 Weasley

-0.68

Ń·

-0.68

wagen

-0.66

okemon

-0.66

asonable

-0.65

POSITIVE LOGITS

ding

1.26

ded

1.15

edded

1.10

embed

1.00

lement

0.89

 embed

0.89

rities

0.79

rained

0.79

 embedded

0.78

iframe

0.77

Activations Density 0.018%

text related to embedding codes within a document

occurrences of the term "embed" and its variations in different contexts

No Comments

No Known Activations