INDEX

Explanations

references to measurements or numeric values associated with data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 behavioural

-0.17

“

-0.15

Behaviour

-0.15

fuck

-0.15

 behaviours

-0.15

 (“

-0.14

 Behaviour

-0.14

-&

-0.14

 signalling

-0.14

 harbour

-0.14

POSITIVE LOGITS

artisan

0.21

 Peru

0.19

 crafts

0.19

 artisan

0.19

 designs

0.18

 artisans

0.18

 craft

0.18

 jewelry

0.18

 Crafts

0.18

 handic

0.17

Activations Density 0.007%

references to measurements or numeric values associated with data

No Comments

No Known Activations