INDEX

Explanations

phrases that emphasize totality or completeness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

afil

-0.14

.AspNet

-0.14

ahan

-0.14

eka

-0.14

neath

-0.14

pb

-0.14

aldi

-0.13

unct

-0.13

opsis

-0.13

enced

-0.13

POSITIVE LOGITS

Ø§Ø¨Øª

0.18

SOLE

0.14

 Zaman

0.14

 Pioneer

0.14

 Hastings

0.13

ÑĶÐ¼

0.13

HELL

0.13

zeigen

0.13

-valu

0.13

Activations Density 0.002%

phrases that emphasize totality or completeness

No Comments

No Known Activations