INDEX

Explanations

first-person singular pronouns and phrases indicating personal experiences or perspectives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eld

-0.16

Smarty

-0.16

 Late

-0.15

å·²

-0.15

 already

-0.15

oup

-0.15

 Bilim

-0.14

dÃŃ

-0.14

odata

-0.14

 handy

-0.13

POSITIVE LOGITS

 originally

0.19

say

0.19

 think

0.18

 initially

0.18

 talk

0.18

 look

0.18

 mention

0.17

look

0.17

 tell

0.17

saw

0.17

Activations Density 0.053%

first-person singular pronouns and phrases indicating personal experiences or perspectives

No Comments

No Known Activations