INDEX

Explanations

phrases relating to events, particularly around loss or mortality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

upply

-0.18

atal

-0.14

DM

-0.14

eczy

-0.14

ÏįÏĢ

-0.13

ummer

-0.13

estruct

-0.13

dm

-0.13

Å¡ÃŃ

-0.13

uture

-0.13

POSITIVE LOGITS

 titular

0.17

 hence

0.17

 Hence

0.15

ikler

0.14

 literally

0.14

iÃ¨res

0.14

Ã¡m

0.14

 preval

0.14

Î¸ÎŃ

0.13

ola

0.13

Activations Density 0.267%

phrases relating to events, particularly around loss or mortality

No Comments

No Known Activations