INDEX

Explanations

terms related to game elements, specifically dice and related concepts

oai_token-act-pair · gpt-3.5-turbo

references to dice and related terminology

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

leased

-0.74

ĺħ

-0.71

ONSORED

-0.70

mbuds

-0.68

artney

-0.66

SUP

-0.66

ppard

-0.66

Kut

-0.65

ovie

-0.65

¿½

-0.65

POSITIVE LOGITS

 dice

0.96

 rolls

0.84

 pool

0.81

 Dice

0.81

eria

0.79

 roll

0.72

pool

0.72

 Cutter

0.69

 Royale

0.68

romeda

0.67

Activations Density 0.011%

terms related to game elements, specifically dice and related concepts

references to dice and related terminology

No Comments

No Known Activations