INDEX

Explanations

phrases related to achievement and recognition

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

Filled

-0.18

 Held

-0.17

Played

-0.16

ãĤ¤ãĥī

-0.16

argon

-0.15

attended

-0.15

istrat

-0.14

 Done

-0.14

/generated

-0.13

.managed

-0.13

POSITIVE LOGITS

 selected

0.44

 chosen

0.39

selected

0.36

 invited

0.35

 given

0.33

 hired

0.31

 picked

0.31

 sent

0.31

chosen

0.30

 asked

0.30

Activations Density 0.461%

phrases related to achievement and recognition

No Comments

No Known Activations