INDEX

Explanations

references to personal identity and experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

Ð±ÑĥÐ´ÑĮ

-0.16

aste

-0.16

odash

-0.15

zos

-0.15

ows

-0.15

do

-0.14

kra

-0.14

ahan

-0.14

uyen

-0.14

_ASSUME

-0.14

POSITIVE LOGITS

 soon

0.21

 likely

0.20

 indeed

0.20

likely

0.19

soon

0.18

 Gilbert

0.16

ekler

0.16

 Likely

0.16

 Soon

0.15

 probably

0.15

Activations Density 0.173%

references to personal identity and experiences

No Comments

No Known Activations