INDEX

Explanations

instances where something is described as impossible

oai_token-act-pair · gpt-3.5-turbo

phrases that express the concept of impossibility

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

avia

-0.88

lance

-0.76

Kings

-0.73

rika

-0.73

lov

-0.72

Pac

-0.72

elle

-0.71

erva

-0.71

ript

-0.71

Reviewed

-0.70

POSITIVE LOGITS

 unnecess

0.81

 feas

0.81

ossible

0.81

 improbable

0.79

 adolesc

0.77

 impossible

0.76

ãĥ¬

0.76

 compr

0.72

 tampering

0.70

istically

0.69

Activations Density 0.019%

instances where something is described as impossible

phrases that express the concept of impossibility

No Comments

No Known Activations