INDEX

Explanations

keywords related to certain actions or procedures within a specific context

oai_token-act-pair · gpt-3.5-turbo

references to various practices, particularly in the context of social and legal issues

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

itialized

-0.86

amaz

-0.72

ilee

-0.72

gin

-0.71

aughter

-0.71

 raspberry

-0.70

arger

-0.70

éĹĺ

-0.69

otos

-0.69

onge

-0.69

POSITIVE LOGITS

 Practices

0.88

 practices

0.85

 practiced

0.78

ually

0.77

itual

0.75

 abuses

0.72

 violations

0.71

 practitioners

0.71

 practice

0.70

 governing

0.69

Activations Density 0.021%

keywords related to certain actions or procedures within a specific context

references to various practices, particularly in the context of social and legal issues

No Comments

No Known Activations