INDEX

Explanations

words and phrases related to attraction or appeal in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

wig

-0.17

pto

-0.16

esta

-0.16

naires

-0.16

áº¯p

-0.15

hips

-0.15

 entrusted

-0.14

PK

-0.14

urre

-0.14

POSITIVE LOGITS

ively

0.33

iveness

0.20

iven

0.17

Ã©gorie

0.16

ingly

0.16

 Ð²Ð½Ð¸Ð¼Ð°Ð½Ð¸Ðµ

0.16

ive

0.16

 nuis

0.15

IVEN

0.15

_mE

0.15

Activations Density 0.033%

words and phrases related to attraction or appeal in various contexts

No Comments

No Known Activations