INDEX

Explanations

references to specific brands and trademarks

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

umat

-0.18

CRET

-0.15

ulin

-0.15

onya

-0.15

Ïħ

-0.15

ode

-0.14

ulen

-0.14

 Ruth

-0.14

ÐºÑĢÐ°

-0.14

ull

-0.14

POSITIVE LOGITS

hem

0.20

Hem

0.20

Yi

0.17

 Perr

0.16

rist

0.16

hem

0.15

ì²Ļ

0.15

onden

0.15

Ã³ng

0.15

 Hemisphere

0.14

Activations Density 0.030%

references to specific brands and trademarks

No Comments

No Known Activations