INDEX

Explanations

phrases or constructs emphasizing the word "of."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

ouch

-0.17

_variance

-0.15

_UNUSED

-0.14

Kad

-0.14

nero

-0.14

ades

-0.14

 meaning

-0.14

urr

-0.13

bert

-0.13

 tick

-0.13

POSITIVE LOGITS

sted

0.17

 sorts

0.17

-course

0.16

ReadWrite

0.15

ingly

0.14

_ASSUME

0.14

ì²Ń

0.14

entimes

0.14

cribe

0.14

 unknown

0.13

Activations Density 0.260%

phrases or constructs emphasizing the word "of."

No Comments

No Known Activations