INDEX

Explanations

references to myths or mythological figures

oai_token-act-pair · gpt-3.5-turbo

references to mythical themes and characters

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

SPONSORED

-0.82

teenth

-0.77

arding

-0.75

ãĤĬ

-0.74

åĤ

-0.74

upt

-0.74

ãĤĭ

-0.73

ä¿

-0.72

aphael

-0.72

ffee

-0.70

POSITIVE LOGITS

 Items

0.83

 Lives

0.82

 Favor

0.81

 Server

0.80

 Face

0.79

 Engine

0.79

 Results

0.78

 Role

0.77

 Thing

0.77

 Fighters

0.77

Activations Density 0.040%

references to myths or mythological figures

references to mythical themes and characters

No Comments

No Known Activations