INDEX

Explanations

negatively framed adverbs or adjectives

oai_token-act-pair · gpt-3.5-turbo

instances of the word "poorly" and related terms describing inadequacy or substandard performance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ership

-0.97

arta

-0.83

dar

-0.82

raltar

-0.80

hip

-0.79

=-=-=-=-=-=-=-=-

-0.76

lean

-0.74

ury

-0.73

leans

-0.72

atorial

-0.69

POSITIVE LOGITS

 suited

0.90

 behaved

0.83

 forgiven

0.72

 misc

0.72

 manag

0.71

 enough

0.71

 dece

0.71

 disadvant

0.70

WARE

0.70

 poorly

0.68

Activations Density 0.007%

negatively framed adverbs or adjectives

instances of the word "poorly" and related terms describing inadequacy or substandard performance

No Comments

No Known Activations