INDEX

Explanations

Java-related terms and keywords

oai_token-act-pair · gpt-3.5-turbo

mentions of the programming language Java and its versions

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uers

-0.79

ablishment

-0.79

pload

-0.76

uer

-0.75

rification

-0.72

kef

-0.71

mble

-0.71

hips

-0.71

ometown

-0.69

POSITIVE LOGITS

 Java

1.22

Java

1.07

EE

0.94

jar

0.92

 Beans

0.84

Script

0.83

SourceFile

0.82

 Clojure

0.81

 java

0.80

Doc

0.80

Activations Density 0.005%

Java-related terms and keywords

mentions of the programming language Java and its versions

No Comments

No Known Activations