INDEX

Explanations

keywords related to logical reasoning and persuasive language

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "arguments."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orporated

-0.67

onet

-0.66

fecture

-0.66

 behold

-0.63

covered

-0.63

ifter

-0.62

 Merrill

-0.62

ishable

-0.62

cko

-0.60

iph

-0.59

POSITIVE LOGITS

 arguments

3.72

 argument

2.66

argument

2.23

 Argument

2.10

Arg

1.71

 objections

1.66

 debates

1.58

 assertions

1.54

Arg

1.52

 arguing

1.49

Activations Density 0.015%

keywords related to logical reasoning and persuasive language

occurrences of the word "arguments."

No Comments

No Known Activations