INDEX

Explanations

specific names and titles within various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

Moo

-0.17

è©

-0.15

mu

-0.14

Mu

-0.14

 Schneider

-0.14

INLINE

-0.13

[/

-0.13

*[

-0.13

mu

-0.13

 Kernel

-0.13

POSITIVE LOGITS

odont

0.17

inia

0.16

 jadx

0.15

iales

0.15

_QMARK

0.15

ecture

0.14

 Ø§ÙĦØ§ØŃ

0.13

Mahon

0.13

rama

0.13

_CONNECTED

0.13

Activations Density 0.021%

specific names and titles within various contexts

No Comments

No Known Activations