INDEX

Explanations

instances of the word 'all' in various forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

-0.18

<context

-0.17

assi

-0.16

éĮ

-0.15

ools

-0.15

-0.14

icz

-0.14

ÃĹ↵↵

-0.14

imas

-0.14

POSITIVE LOGITS

holm

0.19

Ð°ÑģÑĤÐ¸

0.16

 sources

0.16

 getSource

0.16

ande

0.16

ãĥ³ãĥĩ

0.16

 SOUR

0.16

worth

0.15

 gá»ĳc

0.15

.source

0.15

Activations Density 0.004%

instances of the word 'all' in various forms

No Comments

No Known Activations