INDEX

Explanations

actions related to physical interactions with objects

oai_token-act-pair · gpt-3.5-turbo

actions related to physical manipulation or handling

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

âĹ¼

-0.72

journal

-0.71

 hacker

-0.65

akening

-0.63

egal

-0.63

 headlined

-0.63

 Fundamental

-0.63

league

-0.63

Terror

-0.62

Ø§ÙĦ

-0.62

POSITIVE LOGITS

 blades

0.90

 toilets

0.90

 joints

0.89

 nails

0.88

 hairs

0.88

 coils

0.86

 needles

0.85

 jars

0.84

 bowls

0.83

 cones

0.82

Activations Density 0.382%

actions related to physical interactions with objects

actions related to physical manipulation or handling

No Comments

No Known Activations