INDEX

Explanations

phrases related to the setting or location of events

oai_token-act-pair · gpt-3.5-turbo

contexts or settings for stories or narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 actionGroup

-0.79

 dividends

-0.75

 Flavoring

-0.75

 knack

-0.73

 sniff

-0.70

 brakes

-0.69

 enzyme

-0.69

 diligence

-0.68

 privileges

-0.67

 patience

-0.65

POSITIVE LOGITS

 dystopian

1.12

 backdrop

1.02

 fictional

0.98

 dystop

0.95

 circa

0.82

 medieval

0.82

ieval

0.82

 ruins

0.79

 futuristic

0.78

Activations Density 0.198%

phrases related to the setting or location of events

contexts or settings for stories or narratives

No Comments

No Known Activations