INDEX

Explanations

mentions of specific names or identifiers, likely related to people or brands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

-0.18

inspace

-0.15

_contin

-0.15

icode

-0.15

ibName

-0.15

_vue

-0.14

ë¡ľëĤĺ

-0.14

ether

-0.14

FromNib

-0.14

POSITIVE LOGITS

foil

0.17

edom

0.16

stal

0.16

yal

0.15

UNIT

0.14

oman

0.14

azz

0.14

 Troy

0.14

opian

0.14

zb

0.14

Activations Density 0.020%

mentions of specific names or identifiers, likely related to people or brands

No Comments

No Known Activations