INDEX

Explanations

references to authors or writers

oai_token-act-pair · gpt-3.5-turbo

mentions of authorship or attribution in text

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

bia

-0.88

MpServer

-0.80

asy

-0.76

iasm

-0.76

aser

-0.75

pect

-0.74

vous

-0.73

ounter

-0.72

apor

-0.71

ouple

-0.71

POSITIVE LOGITS

 virtue

0.92

 Richard

0.78

 Wizards

0.78

 Warren

0.77

 Hasan

0.77

 Michele

0.74

 Robert

0.74

Hug

0.74

Rod

0.74

 Juan

0.74

Activations Density 0.089%

references to authors or writers

mentions of authorship or attribution in text

No Comments

No Known Activations