INDEX

Explanations

proper nouns that are well-known or recognized for something specific

oai_token-act-pair · gpt-3.5-turbo

phrases that describe someone's recognition or reputation in a particular role or context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

tein

-0.72

 voluntary

-0.65

plet

-0.62

 Guid

-0.59

 evaluation

-0.58

 investigation

-0.58

 randomized

-0.58

wake

-0.58

 possibility

-0.58

istries

-0.57

POSITIVE LOGITS

Offline

0.91

Ø³

0.86

Ú

0.75

ÙĨ

0.73

Ô

0.72

åİ

0.71

ously

0.70

 internationally

0.70

 fond

0.69

Ùħ

0.67

Activations Density 0.056%

proper nouns that are well-known or recognized for something specific

phrases that describe someone's recognition or reputation in a particular role or context

No Comments

No Known Activations