INDEX

Explanations

concepts related to motivation and action-taking processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

TAB

-0.14

emean

-0.14

uppet

-0.13

upiter

-0.13

IDL

-0.13

 quarterly

-0.13

strup

-0.13

upported

-0.13

wol

-0.13

sel

-0.13

POSITIVE LOGITS

 chain

0.28

 Chain

0.23

 started

0.23

chain

0.23

 chains

0.22

chains

0.21

 began

0.21

 casc

0.20

Chain

0.20

 processes

0.20

Activations Density 0.093%

concepts related to motivation and action-taking processes

No Comments

No Known Activations