INDEX

Explanations

terms related to specific items or products

oai_token-act-pair · gpt-3.5-turbo

specific technical terms and jargon related to various topics

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

ciating

-0.84

»Ĵ

-0.76

 referen

-0.73

 Masquerade

-0.70

::::::::

-0.67

idia

-0.66

Ø§ÙĦ

-0.66

ãĥĥãĥī

-0.65

APTER

-0.64

vous

-0.64

POSITIVE LOGITS

auts

0.67

ause

0.65

antis

0.64

ousel

0.64

imus

0.63

ipes

0.63

riers

0.62

ount

0.62

ets

0.62

nets

0.59

Activations Density 0.496%

terms related to specific items or products

specific technical terms and jargon related to various topics

No Comments

No Known Activations