INDEX

Explanations

phrases related to abstract concepts or interpretations, especially with the use of 'sense'

oai_token-act-pair · gpt-3.5-turbo

phrases that refer to different meanings or interpretations of the term "sense."

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

sites

-0.81

avorite

-0.77

astered

-0.76

incinn

-0.72

culosis

-0.69

upiter

-0.66

etsk

-0.66

 downed

-0.65

annis

-0.65

ilight

-0.64

POSITIVE LOGITS

 meaning

0.67

ÙĨ

0.65

forth

0.64

fully

0.64

forward

0.64

ombat

0.63

Ëľ

0.63

 suff

0.63

 thereof

0.62

osis

0.61

Activations Density 0.010%

phrases related to abstract concepts or interpretations, especially with the use of 'sense'

phrases that refer to different meanings or interpretations of the term "sense."

No Comments

No Known Activations