INDEX

Explanations

words related to conquering, conquest, or victory

oai_token-act-pair · gpt-3.5-turbo

terms related to the concept of conquering or victory

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

affer

-0.84

egal

-0.68

 closely

-0.67

unal

-0.66

ensing

-0.66

safety

-0.65

perature

-0.59

alez

-0.59

AE

-0.59

orrow

-0.59

POSITIVE LOGITS

 conquer

0.99

 conquered

0.95

 conquest

0.89

 vanquished

0.84

 conquering

0.83

ãĥĦ

0.81

naire

0.81

ãĥ¼ãĤ¯

0.80

ãĤ¤ãĥĪ

0.79

 conqu

0.77

Activations Density 0.014%

words related to conquering, conquest, or victory

terms related to the concept of conquering or victory

No Comments

No Known Activations