INDEX

Explanations

phrases related to locations or geographic/political entities

oai_token-act-pair · gpt-3.5-turbo

references to geographical locations or entities involved in competitions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

soType

-0.78

ROR

-0.70

 masses

-0.62

oris

-0.60

dayName

-0.60

ppy

-0.59

URL

-0.58

ãģķ

-0.55

cffffcc

-0.54

OCK

-0.54

POSITIVE LOGITS

three

1.62

 three

1.58

 four

1.58

two

1.52

 five

1.45

two

1.43

 seven

1.42

six

1.42

five

1.42

four

1.42

Activations Density 0.511%

phrases related to locations or geographic/political entities

references to geographical locations or entities involved in competitions or events

No Comments

No Known Activations