INDEX

Explanations

descriptions of organizations or companies

oai_token-act-pair · gpt-3.5-turbo

references to organizations or groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 forgetting

-0.75

 spilled

-0.74

 chopping

-0.73

 even

-0.70

 spill

-0.68

 futile

-0.67

 blot

-0.66

 guessing

-0.66

 joking

-0.65

 dangling

-0.64

POSITIVE LOGITS

Overview

1.02

ccording

1.01

Definition

0.97

Fund

0.94

 Purpose

0.93

Generally

0.91

GROUND

0.90

BACK

0.89

Created

0.89

Originally

0.88

Activations Density 0.190%

descriptions of organizations or companies

references to organizations or groups

No Comments

No Known Activations