INDEX

Explanations

phrases related to specific events, legal terms, and authority actions

oai_token-act-pair · gpt-3.5-turbo

conjunctions, particularly the word "and" in various contexts

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

sie

-0.70

yout

-0.69

inem

-0.68

erie

-0.64

quit

-0.63

albeit

-0.62

twitch

-0.62

always

-0.62

huh

-0.59

iltr

-0.58

POSITIVE LOGITS

 other

1.30

 related

1.15

 adjoining

1.12

other

0.99

 Other

0.95

 Related

0.94

 subsequent

0.92

 surrounding

0.92

 elsewhere

0.92

 similar

0.91

Activations Density 0.274%

phrases related to specific events, legal terms, and authority actions

conjunctions, particularly the word "and" in various contexts

No Comments

No Known Activations