INDEX

Explanations

references to specific countries or territories

oai_token-act-pair · gpt-3.5-turbo

references to countries and their respective states or regions

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

tumblr

-0.79

anwhile

-0.76

htaking

-0.69

odynam

-0.68

intensive

-0.67

hern

-0.67

gmail

-0.65

ournals

-0.65

Downloadha

-0.64

odynamic

-0.64

POSITIVE LOGITS

lde

0.72

issance

0.69

isers

0.65

unda

0.65

anchester

0.64

repeat

0.61

lishes

0.60

export

0.60

 withdrew

0.59

 Bundes

0.58

Activations Density 0.252%

references to specific countries or territories

references to countries and their respective states or regions

No Comments

No Known Activations