INDEX

Explanations

terms related to reversing or undoing actions or processes

oai_token-act-pair · gpt-3.5-turbo

terms related to the concept of reverse engineering

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

Interstitial

-1.00

lished

-0.96

utical

-0.82

chens

-0.81

uay

-0.76

rine

-0.76

urated

-0.75

rics

-0.71

chen

-0.70

urate

-0.70

POSITIVE LOGITS

 chronological

1.05

 engineer

0.87

engineering

0.81

balanced

0.76

 engineering

0.75

 symmetry

0.74

 engineered

0.73

 dial

0.71

 causation

0.71

actively

0.70

Activations Density 0.029%

terms related to reversing or undoing actions or processes

terms related to the concept of reverse engineering

No Comments

No Known Activations