INDEX

Explanations

negations or words indicating contradiction

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

ADOR

-0.16

DESC

-0.15

inkel

-0.15

ady

-0.15

hart

-0.14

NotSupportedException

-0.14

å®¿

-0.14

struct

-0.13

Guy

-0.13

POSITIVE LOGITS

ori

0.19

ched

0.18

tingham

0.17

epad

0.17

ches

0.17

abyrin

0.16

aira

0.16

¤

0.15

zsche

0.15

CHK

0.15

Activations Density 0.186%

negations or words indicating contradiction

No Comments

No Known Activations