INDEX

Explanations

references to game performances and outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

igers

-0.16

uro

-0.16

igest

-0.15

/******************************************************************************↵

-0.15

aries

-0.15

ugins

-0.15

 Sang

-0.14

peat

-0.14

iani

-0.14

 Porto

-0.14

POSITIVE LOGITS

lint

0.17

erland

0.15

.Solid

0.15

eri

0.15

ÏħÎº

0.15

onds

0.14

.vertx

0.13

_syntax

0.13

/vue

0.13

idan

0.13

Activations Density 0.061%

references to game performances and outcomes

No Comments

No Known Activations