INDEX

Explanations

terms related to fingers and pointing

oai_token-act-pair · gpt-3.5-turbo

references to fingers and related actions or expressions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Admission

-0.69

Ļ½

-0.66

 Females

-0.64

etheus

-0.62

 Shelby

-0.62

zona

-0.61

zek

-0.60

 Plat

-0.60

 Kuwait

-0.59

 Atmosp

-0.59

POSITIVE LOGITS

prints

1.28

print

1.25

pointer

1.06

 nails

1.02

 pointing

0.96

 pointer

0.91

 fingers

0.91

tip

0.91

 nail

0.90

picking

0.90

Activations Density 0.049%

terms related to fingers and pointing

references to fingers and related actions or expressions

No Comments

No Known Activations