INDEX

Explanations

numbers with percentage values

oai_token-act-pair · gpt-3.5-turbo

occurrences of the article "a" in various contexts

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

agents

-0.70

CHAPTER

-0.67

 extrad

-0.65

 Presents

-0.64

ovie

-0.64

things

-0.63

Edit

-0.62

 dishes

-0.62

authent

-0.61

 outlaw

-0.61

POSITIVE LOGITS

 whopping

1.45

 ratio

1.16

verages

1.05

 staggering

1.04

 median

1.02

 negligible

0.96

 fraction

0.96

 slight

0.94

 dismal

0.92

 margin

0.89

Activations Density 0.155%

numbers with percentage values

occurrences of the article "a" in various contexts

No Comments

No Known Activations