INDEX

Explanations

numerical values, years, and alphanumeric sequences

oai_token-act-pair · gpt-3.5-turbo

numerical data and statistics related to events or achievements

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

Jiu

-0.67

 FedEx

-0.65

 Joshua

-0.63

-0.61

AMA

-0.61

hement

-0.60

 Faul

-0.60

ãĥķãĤ©

-0.58

 Titan

-0.58

Zen

-0.57

POSITIVE LOGITS

ourney

0.79

âĢ²

0.70

itchie

0.67

STON

0.67

WAY

0.65

akedown

0.62

warn

0.60

esides

0.60

 externalToEVAOnly

0.60

')

0.60

Activations Density 0.235%

numerical values, years, and alphanumeric sequences

numerical data and statistics related to events or achievements

No Comments

No Known Activations