INDEX

Explanations

references to letters or written communication

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

orsk

-0.17

emaker

-0.16

umber

-0.15

ot

-0.15

sym

-0.15

oodle

-0.14

esser

-0.14

ë¬¼

-0.14

Unc

-0.14

erala

-0.14

POSITIVE LOGITS

rente

0.16

reuse

0.16

lies

0.15

thew

0.15

ÙĨØ¬

0.15

xeb

0.15

lie

0.15

lob

0.15

ILLISECONDS

0.15

zÅĳ

0.14

Activations Density 0.036%

references to letters or written communication

No Comments

No Known Activations