INDEX

Explanations

negations and expressions of reluctance or uncertainty

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

Ð½Ð³

-0.16

irk

-0.15

inski

-0.14

ams

-0.14

_defaults

-0.14

ÅĦ

-0.14

 defaults

-0.13

onder

-0.13

inis

-0.13

ÐµÑĢÐ¸Ð½

-0.13

POSITIVE LOGITS

cket

0.15

alla

0.14

ctal

0.14

_unused

0.14

HING

0.14

aleigh

0.14

aket

0.13

acket

0.13

à¸²à¸§

0.13

 especial

0.13

Activations Density 0.092%

negations and expressions of reluctance or uncertainty

No Comments

No Known Activations