INDEX

Explanations

references to locations or landmarks, particularly identifiable buildings or addresses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

ÙĦØ§Ùģ

-0.16

Mog

-0.16

hausen

-0.16

 Dash

-0.15

asm

-0.15

oui

-0.15

vais

-0.15

 Halk

-0.15

’Ãł

-0.14

'Ãł

-0.14

POSITIVE LOGITS

iminal

0.16

ense

0.16

0.15

 Carlton

0.15

Esc

0.15

anson

0.15

Ú¯Ø±Ø§ÙĨ

0.14

Univers

0.14

pool

0.14

anus

0.14

Activations Density 0.026%

references to locations or landmarks, particularly identifiable buildings or addresses

No Comments

No Known Activations