INDEX

Explanations

phrases related to education, learning, and training

oai_token-act-pair · gpt-3.5-turbo

references to training and teaching in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

iture

-0.74

 Corker

-0.72

beck

-0.70

 attm

-0.70

endez

-0.66

eca

-0.66

ylum

-0.66

NetMessage

-0.65

 inconven

-0.65

oyer

-0.65

POSITIVE LOGITS

 curriculum

0.92

 basics

0.90

 skills

0.90

 lesson

0.89

 curric

0.86

 rudimentary

0.85

 teach

0.85

 fundamentals

0.84

 lessons

0.83

 Teach

0.83

Activations Density 0.281%

phrases related to education, learning, and training

references to training and teaching in various contexts

No Comments

No Known Activations