INDEX

Explanations

questions or speculations in the form of a statement ending with a question mark

oai_token-act-pair · gpt-3.5-turbo

questions or speculation about events, particularly those involving mystery or uncertainty

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

onomy

-0.85

emn

-0.84

ensable

-0.77

antage

-0.77

orest

-0.75

ordable

-0.75

verages

-0.74

excluding

-0.74

ocre

-0.74

inguished

-0.73

POSITIVE LOGITS

 somewhere

1.24

 somew

1.12

 somehow

1.02

 subconscious

0.95

 somet

0.94

 nefarious

0.91

 something

0.89

 sometime

0.88

 prank

0.86

 intentional

0.85

Activations Density 0.932%

questions or speculations in the form of a statement ending with a question mark

questions or speculation about events, particularly those involving mystery or uncertainty

No Comments

No Known Activations