INDEX

Explanations

comparisons or superlatives related to quantities or degrees

oai_token-act-pair · gpt-3.5-turbo

references to comparative statements involving the term "more."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

istries

-0.77

arta

-0.64

 ancest

-0.63

 defect

-0.62

ĸļ

-0.60

arios

-0.60

 kindred

-0.59

mare

-0.59

 footing

-0.57

ONEY

-0.56

POSITIVE LOGITS

average

0.80

usual

0.76

 slightest

0.74

 average

0.68

 usual

0.67

outer

0.62

ducers

0.62

 others

0.61

ieg

0.61

 actual

0.60

Activations Density 0.209%

comparisons or superlatives related to quantities or degrees

references to comparative statements involving the term "more."

No Comments

No Known Activations