INDEX

Explanations

keywords related to actions or behaviors, especially in the context of criticisms or observations about others' actions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the verb "act" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Teg

-0.66

 Flavoring

-0.65

nic

-0.64

TON

-0.63

burn

-0.62

ickets

-0.61

yip

-0.61

Graphics

-0.60

 Cotton

-0.60

fer

-0.60

POSITIVE LOGITS

uate

1.14

 accordingly

1.05

uated

1.04

 differently

1.04

 decisively

1.02

 responsibly

1.00

 independently

0.98

 aggressively

0.98

 impuls

0.95

 unilaterally

0.94

Activations Density 0.045%

keywords related to actions or behaviors, especially in the context of criticisms or observations about others' actions

occurrences of the verb "act" and its variations

No Comments

No Known Activations