INDEX

Explanations

terms related to specific experiments or technologies named "erm," with different levels of activation suggesting varying degrees of relevance

oai_token-act-pair · gpt-3.5-turbo

references to a specific scientific term or concept

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

vati

-0.72

ched

-0.69

66666666

-0.69

 stake

-0.66

 priced

-0.66

ciating

-0.65

BOOK

-0.63

handled

-0.62

fulness

-0.61

ccording

-0.61

POSITIVE LOGITS

ont

0.92

ilk

0.91

uda

0.90

aceutical

0.88

otor

0.86

ichael

0.85

obile

0.84

ission

0.84

otive

0.83

aintain

0.81

Activations Density 0.018%

terms related to specific experiments or technologies named "erm," with different levels of activation suggesting varying degrees of relevance

references to a specific scientific term or concept

No Comments

No Known Activations