INDEX

Explanations

words related to characteristics, attributes, or descriptors of objects or experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

Rim

-0.15

arton

-0.15

 ambient

-0.15

ØªØ§ÙĨ

-0.14

ancel

-0.14

ason

-0.14

licos

-0.13

Disallow

-0.13

tan

-0.13

Hex

-0.13

POSITIVE LOGITS

ppard

0.16

htar

0.15

idlo

0.15

uffix

0.15

UGE

0.14

bulan

0.14

 NavParams

0.14

dana

0.14

zig

0.14

Ð¾Ð±ÑīÐµ

0.14

Activations Density 0.015%

words related to characteristics, attributes, or descriptors of objects or experiences

No Comments

No Known Activations