INDEX

Explanations

phrases that indicate association or collaboration

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ikel

-0.17

swick

-0.16

Ð´Ð°Ð¼

-0.16

eyer

-0.15

jac

-0.15

iez

-0.14

ouce

-0.14

rak

-0.14

ĶåĽŀ

-0.14

IMARY

-0.14

POSITIVE LOGITS

aleur

0.15

 ways

0.15

illard

0.15

mpz

0.15

iscard

0.15

PIO

0.15

Î®

0.14

 Ways

0.14

trie

0.14

uite

0.14

Activations Density 0.008%

phrases that indicate association or collaboration

No Comments

No Known Activations