INDEX

Explanations

words related to proper nouns and locations

oai_token-act-pair · gpt-3.5-turbo

words related to specific names or titles, particularly focusing on the prefix 'B' and similar patterns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

FORMATION

-0.69

\/\/

-0.67

RFC

-0.65

 Malays

-0.60

FTC

-0.60

xual

-0.59

ĸļ

-0.58

Whereas

-0.57

ources

-0.57

ngth

-0.57

POSITIVE LOGITS

levard

1.11

lehem

1.01

pillar

0.99

apest

0.94

hammad

0.88

rill

0.80

etooth

0.79

abase

0.78

ause

0.78

aneers

0.77

Activations Density 0.126%

words related to proper nouns and locations

words related to specific names or titles, particularly focusing on the prefix 'B' and similar patterns

No Comments

No Known Activations