INDEX

Explanations

Boolean values related to programming logic

oai_token-act-pair · gpt-3.5-turbo

boolean values in coding and programming contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

regor

-0.97

apers

-0.83

agna

-0.83

sels

-0.78

ngth

-0.75

lectic

-0.75

lyak

-0.72

NRS

-0.71

TOR

-0.71

odynam

-0.71

POSITIVE LOGITS

NULL

0.89

 unfocusedRange

0.86

xff

0.79

False

0.78

 msec

0.74

ptr

0.73

()

0.71

 else

0.71

 Boolean

0.69

 Checking

0.67

Activations Density 0.063%

Boolean values related to programming logic

boolean values in coding and programming contexts

No Comments

No Known Activations