INDEX

Explanations

terms related to Gnosticism

oai_token-act-pair · gpt-3.5-turbo

references to Gnostic texts and concepts

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

IFrame

Link

Negative Logits

WHERE

-0.90

LIST

-0.74

Grant

-0.70

ICAN

-0.69

Materials

-0.69

VICE

-0.68

FINE

-0.68

IDENT

-0.67

Condition

-0.67

rawdownloadcloneembedreportprint

-0.66

POSITIVE LOGITS

acles

0.99

ostic

0.96

agons

0.92

omon

0.91

acle

0.89

Gn

0.89

omach

0.88

emonic

0.88

gn

0.88

ollah

0.87

Activations Density 0.011%

terms related to Gnosticism

references to Gnostic texts and concepts

No Known Activations