INDEX

Explanations

phrases or words related to writing or written text

oai_token-act-pair · gpt-3.5-turbo

terms related to scripts and transcription

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

 visitation

-0.71

DERR

-0.70

 Bears

-0.70

 Tomas

-0.67

Geh

-0.66

favorite

-0.64

à©

-0.64

fred

-0.64

Franc

-0.62

 Eclipse

-0.62

POSITIVE LOGITS

osate

1.16

ions

1.05

onite

0.95

ript

0.94

ioned

0.93

ive

0.93

sis

0.92

alid

0.91

entimes

0.90

icut

0.88

Activations Density 0.031%

phrases or words related to writing or written text

terms related to scripts and transcription

No Comments

No Known Activations