INDEX

Explanations

the word "blot."

oai_token-act-pair · gpt-3.5-turbo

instances of the word "blot" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

-1.23

 psychiat

-0.79

luaj

-0.78

pport

-0.76

agra

-0.74

Downloadha

-0.68

ordinate

-0.67

ivism

-0.66

 scrut

-0.65

 cust

-0.63

POSITIVE LOGITS

ks

0.67

lier

0.67

odox

0.66

 Things

0.65

ipple

0.63

 sorts

0.61

 Rodgers

0.60

 STORY

0.60

Won

0.59

FIL

0.58

Activations Density 0.000%

the word "blot."

instances of the word "blot" and its variations

No Comments

No Known Activations