INDEX

Explanations

sentences related to personal experiences and reflections

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 knockout

-0.88

 neglig

-0.81

 tyrann

-0.81

 skelet

-0.78

 metic

-0.78

 enriched

-0.76

 desper

-0.75

 undet

-0.75

 nutshell

-0.75

 endeav

-0.75

POSITIVE LOGITS

Indeed

1.77

Asked

1.75

Others

1.68

Added

1.63

He

1.56

Another

1.55

Still

1.52

Though

1.52

While

1.51

Despite

1.51

Activations Density 0.267%

sentences related to personal experiences and reflections

No Comments

No Known Activations