INDEX

Explanations

words related to flags

oai_token-act-pair · gpt-3.5-turbo

references to flags and flag-related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Argon

-0.72

nesota

-0.72

è£ħ

-0.71

venants

-0.69

 Flavoring

-0.67

ynthesis

-0.65

å¥³

-0.64

axter

-0.62

 subsequ

-0.60

elsen

-0.60

POSITIVE LOGITS

pole

1.56

rant

1.48

ging

1.41

staff

1.23

 bearer

1.09

ged

1.02

stones

0.99

gers

0.97

rants

0.96

rier

0.94

Activations Density 0.047%

words related to flags

references to flags and flag-related concepts

No Comments

No Known Activations