INDEX

Explanations

references to offspring and descendants

oai_token-act-pair · gpt-3.5-turbo

terms related to offspring and their lineage or ancestry

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

idelines

-0.84

neck

-0.75

ogun

-0.72

ewater

-0.71

yer

-0.71

imar

-0.70

asonable

-0.70

redd

-0.68

yers

-0.68

ocal

-0.66

POSITIVE LOGITS

 offspring

1.30

 descendant

0.99

 descendants

0.91

romeda

0.86

 inherit

0.83

 adolesc

0.81

 Emin

0.79

 inher

0.74

Reviewer

0.74

 taxp

0.71

Activations Density 0.020%

references to offspring and descendants

terms related to offspring and their lineage or ancestry

No Comments

No Known Activations