INDEX

Explanations

phrases that express contradictory or negative sentiments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

inity

-0.16

èĽ

-0.16

mdir

-0.16

èĹ

-0.16

RowAt

-0.15

ãĤĩ

-0.15

prene

-0.15

emer

-0.15

ìĬ¬

-0.14

uet

-0.14

POSITIVE LOGITS

 tier

0.16

 Tier

0.15

 twice

0.14

 harder

0.14

ade

0.14

 quant

0.14

tab

0.14

lobal

0.14

Nat

0.14

 pattern

0.14

Activations Density 0.015%

phrases that express contradictory or negative sentiments

No Comments

No Known Activations