INDEX

Explanations

phrases related to repetitions and actions in a given context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

ardash

-0.15

roje

-0.14

otted

-0.14

 INCIDENTAL

-0.14

ownt

-0.13

ivent

-0.13

 Abstract

-0.13

\Abstract

-0.13

marshall

-0.13

ãĥĩãĤ£ãĤ¢

-0.13

POSITIVE LOGITS

 again

0.92

again

0.80

 Again

0.70

Again

0.69

ag

0.61

AGAIN

0.59

_again

0.59

 AGAIN

0.59

 novamente

0.56

 wieder

0.54

Activations Density 0.106%

phrases related to repetitions and actions in a given context

No Comments

No Known Activations