INDEX

Explanations

modal verbs indicating possibility or likelihood

oai_token-act-pair · gpt-3.5-turbo

conditional phrases or expressions of possibility

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

perty

-0.70

 cultivating

-0.68

 Hyder

-0.67

 submitting

-0.66

 performing

-0.64

 inspecting

-0.63

 aiming

-0.63

cius

-0.62

Reviewer

-0.62

 seeking

-0.62

POSITIVE LOGITS

 happen

1.02

 happened

0.99

 happens

0.88

 coincides

0.85

 happ

0.83

 coincided

0.81

 horr

0.80

 translates

0.80

 motivate

0.80

 imply

0.79

Activations Density 0.165%

modal verbs indicating possibility or likelihood

conditional phrases or expressions of possibility

No Comments

No Known Activations