INDEX

Explanations

phrases related to the definition or interpretation of terms

oai_token-act-pair · gpt-3.5-turbo

words and phrases that relate to definitions and meanings

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

 outweigh

-0.68

aldi

-0.66

ÃĥÃĤ

-0.64

vertisement

-0.64

IRO

-0.62

ibling

-0.61

 Skydragon

-0.60

outube

-0.58

 fors

-0.57

 enthusi

-0.57

POSITIVE LOGITS

 term

0.90

 coined

0.82

 acron

0.80

``

0.79

\"

0.79

0.78

 Term

0.77

 definition

0.76

 terminology

0.76

Activations Density 0.484%

phrases related to the definition or interpretation of terms

words and phrases that relate to definitions and meanings

No Comments

No Known Activations