INDEX

Explanations

phrases related to government commitments and regulatory processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

uel

-0.16

mitt

-0.15

Ŀ

-0.15

akit

-0.15

 Bron

-0.15

 gris

-0.15

UEL

-0.15

enda

-0.15

ãĤŃãĥ¥

-0.14

POSITIVE LOGITS

 rope

0.24

 ropes

0.22

 Rope

0.21

 intimate

0.21

rop

0.20

 club

0.19

 chalk

0.19

 regular

0.19

 accord

0.18

 intros

0.17

Activations Density 0.218%

phrases related to government commitments and regulatory processes

No Comments

No Known Activations