INDEX

Explanations

technical terms related to programming and capabilities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

Invoker

-0.16

 yans

-0.15

.addProperty

-0.15

rane

-0.15

BackColor

-0.15

isoft

-0.14

agger

-0.14

 iken

-0.14

peration

-0.14

ember

-0.13

POSITIVE LOGITS

 Manip

0.17

ph

0.17

viewer

0.16

 manip

0.16

 Herald

0.16

_ph

0.16

 manipulate

0.15

Manip

0.15

PH

0.15

PH

0.15

Activations Density 0.030%

technical terms related to programming and capabilities

No Comments

No Known Activations