INDEX

Explanations

numerical values along with specific keywords related to measurements and quantities

oai_token-act-pair · gpt-3.5-turbo

topics and terms associated with energy, investigation, and organizational structures

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ynthesis

-0.81

enance

-0.77

brance

-0.74

urden

-0.73

someone

-0.70

uana

-0.70

cale

-0.70

icism

-0.70

keleton

-0.67

rium

-0.67

POSITIVE LOGITS

 slots

1.05

 categories

1.03

 fronts

1.02

 prototypes

1.01

 locations

1.01

 sessions

1.01

 halves

1.01

 scenarios

1.01

 instances

1.00

 variants

1.00

Activations Density 0.574%

numerical values along with specific keywords related to measurements and quantities

topics and terms associated with energy, investigation, and organizational structures

No Comments

No Known Activations