INDEX

Explanations

numbers related to specific locations or people

oai_token-act-pair · gpt-3.5-turbo

specific names and terms related to locations or people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ifiers

-0.93

gart

-0.88

ians

-0.84

irs

-0.83

ikarp

-0.83

atron

-0.79

arov

-0.79

nas

-0.78

isite

-0.76

icians

-0.76

POSITIVE LOGITS

OPLE

0.83

 Archdemon

0.69

 appre

0.66

 Helena

0.65

 Vide

0.64

 Stability

0.64

Typ

0.64

 freezing

0.62

âĸ¬

0.62

 airborne

0.61

Activations Density 0.014%

numbers related to specific locations or people

specific names and terms related to locations or people

No Comments

No Known Activations