INDEX

Explanations

abbreviations related to universities

oai_token-act-pair · gpt-3.5-turbo

references to unique identification codes or designators

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

 Noir

-0.75

ãĤ©

-0.73

 flares

-0.66

aline

-0.66

 Preferred

-0.65

ãĤ¤ãĥĪ

-0.64

 Emin

-0.63

ãĤ¨ãĥ«

-0.63

ragon

-0.62

 Sloan

-0.62

POSITIVE LOGITS

NA

1.02

JA

0.96

YA

0.94

UGE

0.93

KE

0.93

WT

0.92

CLA

0.92

MA

0.91

PDATED

0.91

BLIC

0.91

Activations Density 0.033%

abbreviations related to universities

references to unique identification codes or designators

No Comments

No Known Activations