INDEX

Explanations

names or entities related to software development or coding

oai_token-act-pair · gpt-3.5-turbo

mentions of specific names and places, particularly focusing on the term "jar."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

SPD

-0.66

ingham

-0.64

äºĶ

-0.63

 sequ

-0.62

 cheapest

-0.59

 Continental

-0.58

ATL

-0.58

 oxid

-0.58

 Elect

-0.56

ATK

-0.56

POSITIVE LOGITS

jar

1.79

Jar

1.13

zan

0.92

rar

0.88

rod

0.83

Jar

0.82

rack

0.82

akia

0.82

olla

0.82

azz

0.82

Activations Density 0.005%

names or entities related to software development or coding

mentions of specific names and places, particularly focusing on the term "jar."

No Comments

No Known Activations