INDEX

Explanations

specific keywords and identifiers commonly used in coding

oai_token-act-pair · gpt-3.5-turbo

instances of programming-related terminology, particularly focusing on user and variable declarations

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

theless

-0.83

zyk

-0.77

tumblr

-0.75

ktop

-0.74

ï¿½ï¿½ï¿½ï¿½

-0.71

âĹ¼

-0.71

CLASSIFIED

-0.71

technical

-0.69

 Newsletter

-0.69

terday

-0.68

POSITIVE LOGITS

0.77

0.76

 init

0.73

0.71

Input

0.68

 args

0.67

():

0.66

ader

0.66

 caps

0.65

()

0.63

Activations Density 0.135%

specific keywords and identifiers commonly used in coding

instances of programming-related terminology, particularly focusing on user and variable declarations

No Comments

No Known Activations