INDEX

Explanations

references to specific people, things, or notable distinctions 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 @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 scare

-0.15

/gcc

-0.15

Ø§Ø¶

-0.14

-UA

-0.14

itou

-0.14

itzer

-0.14

URLRequest

-0.14

munition

-0.13

ActionResult

-0.13

 pressure

-0.13

POSITIVE LOGITS

 caught

0.58

 catch

0.57

 catches

0.56

 catching

0.52

catch

0.50

caught

0.49

 Catch

0.47

Catch

0.45

 Caught

0.44

Caught

0.41

Activations Density 0.192%

references to specific people, things, or notable distinctions in various contexts

No Comments

No Known Activations