INDEX

Explanations

character names or proper nouns

oai_token-act-pair · gpt-3.5-turbo

characters' names and their interactions

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

Reviewer

-0.81

taboola

-0.79

 confir

-0.77

 guiActiveUn

-0.76

Enlarge

-0.75

olphins

-0.74

 EDITION

-0.74

Effective

-0.73

EPA

-0.69

Manufact

-0.69

POSITIVE LOGITS

 Weasley

1.08

 Kenobi

0.95

 Protector

0.88

 Tooth

0.87

 Xander

0.86

 Joker

0.85

 herself

0.85

 Skywalker

0.83

 Knight

0.83

Fey

0.83

Activations Density 0.346%

character names or proper nouns

characters' names and their interactions

No Comments

No Known Activations