INDEX

Explanations

instances of the verb "was" along with related past-tense or passive constructions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

les

-0.17

 cigarettes

-0.15

 conspiracy

-0.15

okers

-0.15

obar

-0.14

.escape

-0.14

aho

-0.14

igure

-0.14

umin

-0.14

atter

-0.14

POSITIVE LOGITS

ÑĢÑĥÐ±

0.14

Ã³wn

0.14

Busy

0.13

Ð»Ð¸ÑĤ

0.13

rypto

0.13

Unsigned

0.13

_MT

0.13

ÑģÐ¾ÑĤ

0.13

athan

0.13

spath

0.13

Activations Density 0.533%

instances of the verb "was" along with related past-tense or passive constructions

No Comments

No Known Activations