INDEX

Explanations

phrases that indicate locations or contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

 ðŁĺī↵↵

-0.15

uges

-0.15

áºł

-0.14

iene

-0.14

allo

-0.14

enk

-0.14

eyJ

-0.14

ibilit

-0.13

ctrine

-0.13

dar

-0.13

POSITIVE LOGITS

akers

0.15

compat

0.15

lla

0.14

 NSUInteger

0.14

angel

0.14

dera

0.14

ERGY

0.14

oke

0.14

ache

0.13

Activations Density 0.178%

phrases that indicate locations or contexts

No Comments

No Known Activations