INDEX

Explanations

phrases that signify an exception or contrast in a sentence

oai_token-act-pair · gpt-3.5-turbo

instances of the word "except."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

natureconservancy

-0.74

pt

-0.56

pires

-0.56

rounder

-0.56

cent

-0.56

hent

-0.55

urity

-0.55

str

-0.54

scr

-0.52

HCR

-0.52

POSITIVE LOGITS

ional

1.42

 insofar

1.05

ing

0.93

ially

0.75

arus

0.74

 maybe

0.73

where

0.71

Â¯

0.70

 perhaps

0.70

ions

0.69

Activations Density 0.036%

phrases that signify an exception or contrast in a sentence

instances of the word "except."

No Comments

No Known Activations