INDEX

Explanations

references to actions that involve acquiring, checking, or interacting with written or digital content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

achts

-0.17

/packages

-0.14

 Kennedy

-0.14

Ð¡Ðŀ

-0.14

.ta

-0.13

Ø±Ø§Ø¯

-0.13

MED

-0.13

/helpers

-0.13

andas

-0.13

Od

-0.13

POSITIVE LOGITS

 tiÃªn

0.25

 ever

0.19

-ever

0.18

 responders

0.17

Ð½Ð°Ñĩ

0.16

-rate

0.15

idar

0.15

èĪĹ

0.15

alf

0.15

ibs

0.15

Activations Density 0.164%

references to actions that involve acquiring, checking, or interacting with written or digital content

No Comments

No Known Activations