INDEX

Explanations

phrases related to communication, consent, and service interactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

isman

-0.16

remaining

-0.16

 Remaining

-0.15

third

-0.15

second

-0.14

elper

-0.14

 only

-0.14

 Only

-0.14

 third

-0.14

 remaining

-0.14

POSITIVE LOGITS

 person

0.30

 particular

0.28

person

0.25

 given

0.25

 GIVEN

0.20

 persons

0.19

 chosen

0.18

 respective

0.18

given

0.18

persons

0.17

Activations Density 0.688%

phrases related to communication, consent, and service interactions

No Comments

No Known Activations