INDEX

Explanations

references to progress and development in 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 @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 ending

-0.16

 ends

-0.15

anke

-0.14

 ended

-0.14

 reduced

-0.14

 transformed

-0.14

 eventual

-0.14

mans

-0.14

POSITIVE LOGITS

 building

0.56

 build

0.51

 builds

0.51

 Building

0.51

building

0.49

Building

0.48

build

0.48

 BUILD

0.47

 Build

0.43

Build

0.43

Activations Density 0.242%

references to progress and development in various contexts

No Comments

No Known Activations

references to progress and development in various contexts

No Comments

No Known Activations