INDEX

Explanations

phrases indicating high and low quantities in comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

-0.16

anny

-0.16

argo

-0.16

á»Ĳ

-0.15

ULO

-0.15

_fmt

-0.14

ren

-0.14

ward

-0.13

UX

-0.13

aq

-0.13

POSITIVE LOGITS

ovi

0.15

lest

0.14

borough

0.14

bounce

0.14

roken

0.14

licken

0.14

ÏģÎ¯

0.14

Nar

0.13

=sum

0.13

åĩĮ

0.13

Activations Density 0.021%

phrases indicating high and low quantities in comparisons

No Comments

No Known Activations