INDEX

Explanations

references to the name "Benedict."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

çĳŁ

-0.15

tron

-0.15

zap

-0.15

 pupper

-0.15

Ð°Ð»ÐµÐ¶

-0.15

ÑĩÐºÐ¸

-0.14

erer

-0.14

ãĥ³ãĥĸ

-0.14

sey

-0.14

Ø¯ÙĨ

-0.14

POSITIVE LOGITS

ifact

0.17

amins

0.16

etto

0.16

ict

0.15

ue

0.15

Ven

0.15

empt

0.15

IRST

0.15

itas

0.15

amus

0.15

Activations Density 0.008%

references to the name "Benedict."

No Comments

No Known Activations