INDEX

Explanations

conditional phrases and past participles indicating actions related to policies or regulations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

JsonValue

-0.16

AGMA

-0.15

ãĥ«ãĥĪ

-0.15

deen

-0.15

adt

-0.14

clc

-0.14

à¹Ħà¸§

-0.14

 ÙħÙĨØªØ´Ø±

-0.14

atoria

-0.14

astos

-0.13

POSITIVE LOGITS

 asked

0.58

 Asked

0.47

Asked

0.46

 told

0.44

 given

0.36

ask

0.36

 requested

0.35

 urged

0.31

 instructed

0.30

 contacted

0.29

Activations Density 0.165%

conditional phrases and past participles indicating actions related to policies or regulations

No Comments

No Known Activations