INDEX

Explanations

phrases and attitudes relating to personal experiences and advice

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 posit

-0.17

"..

-0.16

 prag

-0.15

 commission

-0.15

posit

-0.14

 prive

-0.14

 deflate

-0.14

 fashioned

-0.14

Ð»Ð¸ÑĪ

-0.14

 similarly

-0.14

POSITIVE LOGITS

get

0.18

wayne

0.18

èĭ

0.18

 figure

0.16

 stay

0.15

get

0.15

exceptions

0.14

pee

0.14

 know

0.14

 gets

0.14

Activations Density 0.653%

phrases and attitudes relating to personal experiences and advice

No Comments

No Known Activations