INDEX

Explanations

names, particularly those with "amar" and "ar"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

kili

-0.16

emachine

-0.16

emi

-0.15

 æĺ¥

-0.15

Æ°Æ¡ng

-0.15

voke

-0.14

enson

-0.14

emic

-0.14

STRU

-0.14

POSITIVE LOGITS

gence

0.17

Ú©Ø²

0.17

ican

0.16

chants

0.16

quis

0.15

ref

0.15

itus

0.15

ullo

0.14

omor

0.14

ãĥ¼ãĥ³

0.14

Activations Density 0.030%

names, particularly those with "amar" and "ar"

No Comments

No Known Activations