INDEX

Explanations

information technology terms and phrases

oai_token-act-pair · gpt-3.5-turbo

indicators of specific actions or activities involving settings and parameters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

xual

-0.86

ĸļ

-0.86

 doomed

-0.82

 squee

-0.80

 spir

-0.78

 unexplained

-0.77

 dece

-0.76

 grappling

-0.75

 foreskin

-0.75

 grips

-0.74

POSITIVE LOGITS

Additionally

1.63

However

1.62

Similarly

1.57

Also

1.56

Other

1.56

Another

1.54

Both

1.54

Meanwhile

1.53

Furthermore

1.52

Others

1.51

Activations Density 0.543%

information technology terms and phrases

indicators of specific actions or activities involving settings and parameters

No Comments

No Known Activations