INDEX

Explanations

specific terms related to preventing or stopping negative events or actions

oai_token-act-pair · gpt-3.5-turbo

phrases emphasizing prevention and avoidance of negative events or situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 largeDownload

-0.89

soDeliveryDate

-0.87

REL

-0.75

å¦

-0.68

é¾įåĸļå£«

-0.66

çľ

-0.64

ettle

-0.64

 congr

-0.63

 Cinem

-0.62

 unfocusedRange

-0.62

POSITIVE LOGITS

 harm

0.91

 congestion

0.85

 harmful

0.81

 accidental

0.80

 costly

0.80

 unwanted

0.80

 unnecessary

0.78

 deterioration

0.78

 misuse

0.78

 pitfalls

0.78

Activations Density 0.179%

specific terms related to preventing or stopping negative events or actions

phrases emphasizing prevention and avoidance of negative events or situations

No Comments

No Known Activations