INDEX

Explanations

references to unique or particular entities or characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

Eg

-0.17

uchs

-0.15

onica

-0.15

Cour

-0.15

uki

-0.14

ika

-0.14

Î»Î¹Î¬

-0.14

à¥ĭà¤§

-0.13

makt

-0.13

enes

-0.13

POSITIVE LOGITS

 nonetheless

0.15

 nevertheless

0.14

ÅĻet

0.14

éĢı

0.14

eton

0.14

antal

0.14

DSA

0.14

eral

0.13

istine

0.13

-dismiss

0.13

Activations Density 0.176%

references to unique or particular entities or characteristics

No Comments

No Known Activations