INDEX

Explanations

mentions or references to a specific place or person named "Or"

oai_token-act-pair · gpt-3.5-turbo

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

DERR

-0.65

 stripes

-0.59

 bases

-0.58

 unions

-0.56

 profiling

-0.56

 conver

-0.55

 deduction

-0.54

BILITIES

-0.54

 floating

-0.53

 traps

-0.53

POSITIVE LOGITS

chard

1.47

chid

1.44

phan

1.42

phans

1.42

pheus

1.37

lando

1.36

leans

1.31

anges

1.29

thodox

1.29

acles

1.23

Activations Density 0.023%

mentions or references to a specific place or person named "Or"

No Comments

No Known Activations