INDEX

Explanations

numerals and their ordinal representations in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

hoff

-0.19

 Apost

-0.14

mou

-0.14

ossal

-0.14

 Criterion

-0.14

 skill

-0.13

-fi

-0.13

LINE

-0.13

.chain

-0.13

hol

-0.13

POSITIVE LOGITS

final

0.16

indre

0.15

 final

0.15

.rd

0.14

.final

0.14

oler

0.14

iese

0.14

 Bard

0.14

_ev

0.14

_final

0.14

Activations Density 0.090%

numerals and their ordinal representations in text

No Comments

No Known Activations

numerals and their ordinal representations in text

No Comments

No Known Activations