INDEX

Explanations

comparative words or phrases, indicating a higher degree or level of something

oai_token-act-pair · gpt-3.5-turbo

the word "even," indicating an emphasis on intensity or degree

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

emp

-0.78

ahime

-0.73

ãĥ¼ãĥ³

-0.72

onomous

-0.70

rend

-0.70

ãĤ´ãĥ³

-0.69

utra

-0.69

ected

-0.66

anu

-0.65

ĺħ

-0.65

POSITIVE LOGITS

handed

1.24

 worse

1.07

 tighter

1.06

 tougher

1.04

 sharper

1.04

 stricter

1.03

 harsher

1.03

 harder

1.02

 stronger

1.02

 bigger

1.02

Activations Density 0.046%

comparative words or phrases, indicating a higher degree or level of something

the word "even," indicating an emphasis on intensity or degree

No Comments

No Known Activations