INDEX

Explanations

references to memory and organizational structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

andes

-0.15

ASI

-0.15

atcher

-0.14

harma

-0.14

iams

-0.14

elmet

-0.14

èĻ

-0.14

plr

-0.14

 Bale

-0.14

ubat

-0.14

POSITIVE LOGITS

its

0.18

å®ĥä»¬

0.15

Its

0.15

uch

0.15

nung

0.14

lingen

0.14

0.13

ë§Ŀ

0.13

fro

0.13

Activations Density 0.436%

references to memory and organizational structures

No Comments

No Known Activations