INDEX

Explanations

products or items described as free of certain ingredients or components

oai_token-act-pair · gpt-3.5-turbo

connections between various concepts, especially highlighting open or free elements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

arya

-0.72

Changes

-0.70

jong

-0.68

 Levels

-0.67

 Factors

-0.66

aurus

-0.66

icators

-0.64

arts

-0.64

ipes

-0.64

tle

-0.63

POSITIVE LOGITS

 insured

1.06

 unregulated

1.05

 unrestricted

1.04

 transparent

1.02

 unden

1.01

 uninterrupted

0.98

 uninsured

0.97

 reversible

0.96

 nonpartisan

0.96

 encrypted

0.93

Activations Density 0.289%

products or items described as free of certain ingredients or components

connections between various concepts, especially highlighting open or free elements

No Comments

No Known Activations