INDEX

Explanations

references to journalism and reporting-related activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

å¿

-0.14

 Î´Îµ

-0.13

esson

-0.13

ellen

-0.13

fen

-0.13

Ð¾ÐºÐ¾Ð½

-0.13

Ã´

-0.13

ç»ĵ

-0.13

_fragment

-0.13

ductor

-0.12

POSITIVE LOGITS

 covering

0.82

 coverage

0.80

 cover

0.74

 covered

0.72

 Coverage

0.71

 Cover

0.69

coverage

0.68

covering

0.68

 covers

0.68

Cover

0.67

Activations Density 0.186%

references to journalism and reporting-related activities

No Comments

No Known Activations