INDEX

Explanations

phrases indicating opposition or disagreement

oai_token-act-pair · gpt-3.5-turbo

references to policies or legislative measures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

elaide

-0.77

 suppose

-0.74

--+

-0.74

 namely

-0.74

replace

-0.72

 besides

-0.71

 wherein

-0.70

een

-0.69

NB

-0.68

ngth

-0.67

POSITIVE LOGITS

 notion

1.39

 idea

1.37

 possibility

1.16

 latter

1.11

 validity

1.10

 outcome

1.09

 entirety

1.08

 legality

1.06

 merits

1.03

 constitution

1.02

Activations Density 0.192%

phrases indicating opposition or disagreement

references to policies or legislative measures

No Comments

No Known Activations