INDEX

Explanations

commands and technical instructions for software development

oai_token-act-pair · gpt-3.5-turbo

negative indicators or commands in a programming context

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

ierrez

-0.77

 Norris

-0.75

 Rounds

-0.72

 Jasper

-0.67

 Gunn

-0.65

Sik

-0.65

 Lucas

-0.64

 Horowitz

-0.64

 Ukrain

-0.64

 interviewing

-0.64

POSITIVE LOGITS

sama

1.13

generic

1.13

level

1.03

type

1.02

utils

1.01

style

1.00

based

1.00

wrapper

1.00

related

0.98

san

0.98

Activations Density 0.090%

commands and technical instructions for software development

negative indicators or commands in a programming context

No Comments

No Known Activations