INDEX

Explanations

occurrences of specific numerical values or sentences that mention numbers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

ekil

-0.18

ãĤ¶ãĥ¼

-0.15

odyn

-0.15

uffer

-0.14

ague

-0.14

awn

-0.14

anic

-0.14

ãĤ¹ãĤ«

-0.14

awner

-0.14

ynos

-0.14

POSITIVE LOGITS

ale

0.16

inkle

0.16

mites

0.15

ichen

0.15

ables

0.14

rous

0.14

ival

0.14

eval

0.14

Ð°ÑģÑģ

0.14

 rare

0.13

Activations Density 0.056%

occurrences of specific numerical values or sentences that mention numbers

No Comments

No Known Activations