INDEX

Explanations

technical or numerical terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

aben

-0.18

uste

-0.16

likler

-0.15

SCREEN

-0.15

zing

-0.15

Ð»Ð°ÑĢÐ°

-0.15

_broadcast

-0.15

usta

-0.14

âĨĲ

-0.14

 Output

-0.14

POSITIVE LOGITS

boru

0.25

ingu

0.20

ÑģÑĤÐ²Ñĥ

0.20

ÑĥÐ¼Ñĥ

0.20

usu

0.19

odu

0.19

Ð¾ÑĢÑĥ

0.19

Ð°ÑĤÐºÑĥ

0.18

 ÑģÐ°Ð¼Ð¾Ð¼

0.18

 hotelu

0.18

Activations Density 0.007%

technical or numerical terminology

No Comments

No Known Activations