INDEX

Explanations

phrases or sentences containing subjective evaluations or opinions

oai_token-act-pair · gpt-3.5-turbo

statements that affirm or negate a particular idea or fact

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ESE

-0.72

çīĪ

-0.72

igraph

-0.69

 Attend

-0.69

ULE

-0.68

éĹĺ

-0.68

iphate

-0.67

ĸļ

-0.66

anwhile

-0.66

 Unle

-0.65

POSITIVE LOGITS

 understandable

1.39

 conceivable

1.36

 plausible

1.33

 interesting

1.33

 doubtful

1.30

 unclear

1.25

 unfortunate

1.24

 conjecture

1.23

 untrue

1.18

 irrelevant

1.18

Activations Density 0.181%

phrases or sentences containing subjective evaluations or opinions

statements that affirm or negate a particular idea or fact

No Comments

No Known Activations