INDEX

Explanations

proper nouns representing individuals' names

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "kan."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Clintons

-0.76

IDENT

-0.70

 Thumbnails

-0.69

Grade

-0.67

Ö¼

-0.63

âĸĵ

-0.63

 predictive

-0.62

ODUCT

-0.62

 PROG

-0.61

 gracious

-0.61

POSITIVE LOGITS

istani

1.10

enei

1.01

kan

0.99

thal

0.96

adian

0.95

amel

0.93

kered

0.91

pai

0.91

ikan

0.90

wi

0.89

Activations Density 0.004%

proper nouns representing individuals' names

occurrences of the word "kan."

No Comments

No Known Activations