INDEX

Explanations

words related to geographical locations or proper nouns

oai_token-act-pair · gpt-3.5-turbo

suffixes or affixes in words, particularly common endings and modifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Bridgewater

-0.66

ï¸ı

-0.63

Yen

-0.60

 Emin

-0.58

NPR

-0.58

 buds

-0.55

 Bloomberg

-0.55

 Petraeus

-0.54

Secure

-0.54

 Pastebin

-0.54

POSITIVE LOGITS

phia

0.89

/-

0.80

fw

0.74

Magikarp

0.74

kj

0.72

gian

0.72

fter

0.71

hod

0.70

ansion

0.69

ocious

0.69

Activations Density 0.405%

words related to geographical locations or proper nouns

suffixes or affixes in words, particularly common endings and modifications

No Comments

No Known Activations