INDEX

Explanations

conjunctions and connecting phrases that indicate relationships or associations between ideas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

iaux

-0.17

ella

-0.17

egen

-0.16

 Pump

-0.15

terra

-0.15

 punch

-0.15

geme

-0.14

lag

-0.14

 Unified

-0.14

argins

-0.14

POSITIVE LOGITS

sher

0.15

sto

0.15

ilim

0.15

ubre

0.14

ogg

0.14

WindowState

0.14

é«

0.14

upal

0.14

Î¼Î±Î¹

0.14

ãĥ¼ãĥĨãĤ£

0.14

Activations Density 1.107%

conjunctions and connecting phrases that indicate relationships or associations between ideas

No Comments

No Known Activations