INDEX

Explanations

occurrences of the word 'ie'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

è´´

-0.16

undi

-0.14

estroy

-0.14

Caption

-0.14

ocket

-0.14

erg

-0.14

esz

-0.14

 Spar

-0.14

cene

-0.14

illet

-0.14

POSITIVE LOGITS

anzi

0.17

wick

0.16

ering

0.16

umu

0.15

çķ

0.15

wiÄħ

0.15

izoph

0.15

 Atlas

0.14

essen

0.14

 nerv

0.14

Activations Density 0.001%

occurrences of the word 'ie'

No Comments

No Known Activations