INDEX

Explanations

negations and auxiliary verbs in the context of composition or writing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

comb

-0.15

urgy

-0.15

Ð±Ð¸

-0.14

íĺĳ

-0.14

 Plains

-0.14

ktop

-0.14

&C

-0.14

_exc

-0.13

PECT

-0.13

erosis

-0.13

POSITIVE LOGITS

meer

0.18

DOI

0.15

oes

0.15

 debut

0.14

abi

0.14

yyn

0.14

 Fight

0.14

 Pruitt

0.14

ovo

0.14

 contingent

0.14

Activations Density 0.004%

negations and auxiliary verbs in the context of composition or writing

No Comments

No Known Activations