INDEX

Explanations

acronyms or specific terms related to military or defense systems

oai_token-act-pair · gpt-3.5-turbo

references to a specific air defense system

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ãĤ¡

-0.82

 pains

-0.72

 shorth

-0.67

ĸļ

-0.64

Ń·

-0.64

 backwards

-0.64

ãĤ©

-0.62

 Canaver

-0.61

Fey

-0.60

 Borders

-0.60

POSITIVE LOGITS

venture

1.22

vance

1.12

VERTIS

1.12

ventures

1.05

AD

1.02

VERTISEMENT

0.97

olph

0.96

illac

0.95

DL

0.94

APTER

0.93

Activations Density 0.010%

acronyms or specific terms related to military or defense systems

references to a specific air defense system

No Comments

No Known Activations