INDEX

Explanations

words related to economic concepts, particularly related to financial statements like "net" and "income"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "net" 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 @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Chrys

-0.82

 Titus

-0.74

edited

-0.73

dated

-0.72

 Claud

-0.71

 Babe

-0.68

CHAPTER

-0.65

dates

-0.64

 Doctors

-0.64

 Mania

-0.64

POSITIVE LOGITS

net

4.02

Net

2.47

 nets

2.44

net

2.26

Net

2.01

NET

2.00

NET

1.57

nets

1.52

 netted

1.50

web

1.13

Activations Density 0.017%

words related to economic concepts, particularly related to financial statements like "net" and "income"

occurrences of the word "net" in various contexts

No Comments

No Known Activations