INDEX

Explanations

words related to terrorist attacks

oai_token-act-pair · gpt-3.5-turbo

references to terrorist attacks

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

dit

-0.97

agin

-0.65

oba

-0.64

 mineral

-0.64

utical

-0.64

ãĤ©

-0.63

theless

-0.62

aple

-0.62

}}}

-0.61

idge

-0.61

POSITIVE LOGITS

 perpetrated

1.01

 against

1.01

 spree

0.99

abad

0.90

 attacks

0.86

 mastermind

0.86

ivity

0.85

CVE

0.84

 retaliation

0.83

 rampage

0.82

Activations Density 0.050%

words related to terrorist attacks

references to terrorist attacks

No Comments

No Known Activations