INDEX

Explanations

words related to real-life events or biographies

oai_token-act-pair · gpt-3.5-turbo

references to real-life experiences and narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

sonian

-0.86

agger

-0.81

rosc

-0.77

 Flavoring

-0.77

mented

-0.77

ractive

-0.75

oug

-0.73

Wed

-0.72

Jaw

-0.71

atchewan

-0.69

POSITIVE LOGITS

 expectancy

1.02

guards

0.92

blood

0.85

liness

0.77

 experien

0.76

ogue

0.76

boats

0.76

guard

0.75

 experiences

0.73

forms

0.73

Activations Density 0.033%

words related to real-life events or biographies

references to real-life experiences and narratives

No Comments

No Known Activations