INDEX

Explanations

specific references to chess and competitive contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

sac

-0.15

uja

-0.15

Sac

-0.15

Sac

-0.14

oney

-0.14

apers

-0.14

info

-0.14

iterr

-0.14

Hak

-0.14

Ch

-0.14

POSITIVE LOGITS

ÏģÎ±Î½

0.16

elay

0.15

.synthetic

0.15

THR

0.15

TEX

0.15

ABI

0.14

etas

0.14

-delay

0.14

Broadcast

0.14

atching

0.14

Activations Density 0.041%

specific references to chess and competitive contexts

No Comments

No Known Activations