INDEX

Explanations

proper names in text

oai_token-act-pair · gpt-3.5-turbo

alphanumeric sequences and abbreviations related to names and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 additive

-0.68

 prol

-0.64

 mosaic

-0.64

IUM

-0.62

 brigade

-0.62

 VIDEOS

-0.62

 conditional

-0.61

 samples

-0.61

 commute

-0.60

 overboard

-0.59

POSITIVE LOGITS

midt

1.04

Ã¤

1.00

Ã¼

0.97

itz

0.93

orthy

0.89

eker

0.89

utter

0.89

orde

0.89

Ã¶r

0.88

acker

0.88

Activations Density 0.156%

proper names in text

alphanumeric sequences and abbreviations related to names and titles

No Comments

No Known Activations