INDEX

Explanations

symbols, punctuation, and formatting related to coding or markup language

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

asts

-0.15

vez

-0.14

mav

-0.13

ÑģÐ¾ÑĢ

-0.13

ä¹ĥ

-0.13

 Peel

-0.13

Ð½Ð¸Ð²ÐµÑĢ

-0.13

verbosity

-0.13

chalk

-0.13

ems

-0.12

POSITIVE LOGITS

Ã¤Ã¤n

0.16

lez

0.14

ANNEL

0.14

ãĥ¼ãĥģ

0.14

0.13

alam

0.13

eden

0.13

lam

0.13

Activations Density 0.063%

symbols, punctuation, and formatting related to coding or markup language

No Comments

No Known Activations