INDEX

Explanations

instances where the word "maybe" is followed by a statement

oai_token-act-pair · gpt-3.5-turbo

instances of the word "it" and its various contexts and forms

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

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

ãģ®éŃĶ

-0.77

ailable

-0.69

ulner

-0.65

verning

-0.64

aii

-0.63

ament

-0.63

 è£ıç

-0.63

 purportedly

-0.63

 extensively

-0.62

licts

-0.62

POSITIVE LOGITS

 someday

1.07

 underest

1.06

 wiser

0.95

 underestimated

0.93

 overest

0.91

 coincidence

0.89

 underrated

0.87

 subconscious

0.87

 misunder

0.86

 somew

0.85

Activations Density 0.245%

instances where the word "maybe" is followed by a statement

instances of the word "it" and its various contexts and forms

No Comments

No Known Activations