INDEX

Explanations

references to military ranks and titles, particularly "Admiral."

oai_token-act-pair · gpt-3.5-turbo

words related to naval ranks and context

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ramid

-0.73

chin

-0.71

 Kavanaugh

-0.68

tics

-0.67

THER

-0.64

gm

-0.63

ters

-0.63

turn

-0.62

medium

-0.62

profit

-0.62

POSITIVE LOGITS

adm

1.09

ittance

1.06

ixture

0.92

 Admir

0.91

iral

0.91

ably

0.88

opol

0.86

icion

0.86

ixed

0.85

abilia

0.82

Activations Density 0.006%

references to military ranks and titles, particularly "Admiral."

words related to naval ranks and context

No Comments

No Known Activations