INDEX

Explanations

character strings that signify structured data or messages, particularly in code or programming contexts

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

ĥn

-0.15

eyse

-0.15

zik

-0.15

@js

-0.15

à¸±à¸ķ

-0.15

ovel

-0.14

ä¸Ģé¡µ

-0.14

.SIG

-0.14

folio

-0.14

POSITIVE LOGITS

coni

0.16

erre

0.15

illez

0.15

ibri

0.15

acht

0.15

 Bowman

0.14

reek

0.14

roit

0.14

cleanup

0.14

 Luca

0.14

Activations Density 0.041%

character strings that signify structured data or messages, particularly in code or programming contexts

No Comments

No Known Activations