INDEX

Explanations

long number sequences

oai_token-act-pair · gpt-3.5-turbo

numerical data and measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

istry

-0.64

 scenery

-0.64

 Lama

-0.60

sed

-0.60

ressor

-0.59

 Kurds

-0.59

 Vide

-0.59

 Berks

-0.58

ween

-0.57

 Spiel

-0.57

POSITIVE LOGITS

0.90

0.89

0.88

0.87

0.86

0.85

Activations Density 0.124%

long number sequences

numerical data and measurements

No Comments

No Known Activations