INDEX

Explanations

words related to research findings and outcomes

oai_token-act-pair · gpt-3.5-turbo

references to results from studies or assessments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Roots

-0.81

agine

-0.75

resent

-0.66

 Passage

-0.65

ategory

-0.63

afort

-0.62

 actionGroup

-0.61

pton

-0.60

Pag

-0.59

yu

-0.59

POSITIVE LOGITS

 obtained

0.89

aries

0.88

iveness

0.84

ively

0.84

 thereof

0.83

ivity

0.80

 results

0.74

 achieved

0.71

results

0.68

aldehyde

0.66

Activations Density 0.063%

words related to research findings and outcomes

references to results from studies or assessments

No Comments

No Known Activations