INDEX

Explanations

references to numbers or numerical concepts

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

New Auto-Interp

Configuration

fnlp/Llama3_1-8B-Base-LXM-8x/Llama3_1-8B-Base-L22M-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.hook_mlp_out

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

StackNavigator

-0.17

ogl

-0.17

ables

-0.16

eva

-0.14

ger

-0.14

sten

-0.14

_hat

-0.14

 hashed

-0.14

 Balance

-0.13

POSITIVE LOGITS

enci

0.17

iggs

0.15

olum

0.14

èįī

0.14

igsaw

0.14

eni

0.14

Ð´Ð¾Ð²

0.14

QUARE

0.14

piration

0.14

atha

0.14

Activations Density 0.030%

references to numbers or numerical concepts

No Comments

No Known Activations