INDEX

Explanations

conditional statements and boolean logic

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 thÃŃ

-0.17

reau

-0.15

ertest

-0.15

Ã¦kke

-0.15

rikes

-0.15

_vlog

-0.15

ieux

-0.14

StackNavigator

-0.14

izu

-0.14

hiba

-0.14

POSITIVE LOGITS

=YES

0.17

 TRUE

0.16

True

0.16

=True

0.16

amon

0.15

ãĥ³ãĥĲ

0.15

 Anita

0.15

0.14

 True

0.14

Activations Density 0.053%

conditional statements and boolean logic

No Comments

No Known Activations