INDEX

Explanations

semicolons (;)

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

ird

-0.17

oyer

-0.15

@}

-0.14

fty

-0.14

orny

-0.14

ickt

-0.14

isas

-0.14

Nin

-0.14

veau

-0.14

eller

-0.14

POSITIVE LOGITS

/Area

0.16

metatable

0.15

atore

0.14

âĻª↵↵

0.14

atoi

0.14

ÏĩÏģÎ¹

0.14

0.13

adora

0.13

ERRU

0.13

iali

0.13

Activations Density 0.008%

semicolons (;)

No Comments

No Known Activations