INDEX

Explanations

comparative phrases highlighting differences or similarities

oai_token-act-pair · gpt-3.5-turbo

instances of comparisons between different subjects or entities

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

authorized

-0.72

 Polo

-0.64

dar

-0.61

aida

-0.58

STAR

-0.58

gren

-0.56

 Semin

-0.55

affe

-0.54

primary

-0.54

acket

-0.54

POSITIVE LOGITS

to

1.05

 thereto

1.01

 favorably

0.97

 favour

0.82

 unto

0.79

 with

0.76

to

0.72

icut

0.71

lihood

0.67

 compared

0.67

Activations Density 0.028%

comparative phrases highlighting differences or similarities

instances of comparisons between different subjects or entities

No Comments

No Known Activations