INDEX

Explanations

references to courthouses and related judicial terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

ores

-0.21

-0.15

emean

-0.15

ei

-0.15

endi

-0.15

evi

-0.15

pher

-0.14

ex

-0.14

unb

-0.14

pe

-0.14

POSITIVE LOGITS

tesy

0.25

tyard

0.24

ousel

0.21

iers

0.20

ney

0.19

rier

0.18

riel

0.18

thouse

0.18

onne

0.17

atatype

0.17

Activations Density 0.006%

references to courthouses and related judicial terminology

No Comments

No Known Activations