INDEX

Explanations

adjectives related to characteristics or qualities

oai_token-act-pair · gpt-3.5-turbo

adjectives and descriptors related to size, strength, and quality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

etically

-0.65

 Pupp

-0.64

 Commissioner

-0.62

 Nest

-0.62

 Patriarch

-0.62

ably

-0.60

 Doct

-0.60

 Garrison

-0.59

appropriately

-0.59

 Rabb

-0.59

POSITIVE LOGITS

cellaneous

0.74

 carbohyd

0.72

frames

0.69

arms

0.68

 subur

0.67

sea

0.67

0.66

wear

0.66

harm

0.66

wave

0.66

Activations Density 0.288%

adjectives related to characteristics or qualities

adjectives and descriptors related to size, strength, and quality

No Comments

No Known Activations