INDEX

Explanations

specific nouns related to various people or objects

oai_token-act-pair · gpt-3.5-turbo

references to groups of people and their circumstances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Rebellion

-0.65

TPPStreamerBot

-0.64

 Colour

-0.64

owan

-0.61

 Gloss

-0.59

 Bows

-0.58

cellence

-0.57

 Trafford

-0.56

 Redemption

-0.56

 Defenders

-0.56

POSITIVE LOGITS

 consisted

1.08

 resulted

0.80

 succeeded

0.77

 reportedly

0.76

 amounted

0.75

atto

0.73

 lured

0.72

 belonged

0.72

 culminated

0.71

 included

0.70

Activations Density 0.888%

specific nouns related to various people or objects

references to groups of people and their circumstances

No Comments

No Known Activations