INDEX

Explanations

statistical or comparative statements

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate probabilities or likelihoods related to various subjects

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

zeb

-0.84

ilts

-0.79

inth

-0.76

ornia

-0.75

uart

-0.74

adel

-0.73

ylan

-0.72

ighth

-0.72

inelli

-0.72

andan

-0.71

POSITIVE LOGITS

 than

0.78

 prey

0.76

Øª

0.71

 likely

0.69

âĢł

0.69

 infer

0.68

 demographics

0.68

 divorce

0.67

 responders

0.66

 opting

0.66

Activations Density 0.028%

statistical or comparative statements

phrases that indicate probabilities or likelihoods related to various subjects

No Comments

No Known Activations