INDEX

Explanations

phrases indicating preparedness or anticipation of future events

oai_token-act-pair · gpt-3.5-turbo

conditional phrases and scenarios involving potential events or emergencies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Harold

-0.69

thank

-0.68

riet

-0.68

 Flavoring

-0.67

 Darling

-0.65

avery

-0.63

Alb

-0.62

 Allan

-0.62

 Theodore

-0.62

ertodd

-0.62

POSITIVE LOGITS

 lapse

0.82

 unforeseen

0.82

 happen

0.75

 slightest

0.73

 breach

0.72

 sufficiently

0.71

mble

0.70

 circumstances

0.70

 circumstance

0.70

 conflict

0.69

Activations Density 0.186%

phrases indicating preparedness or anticipation of future events

conditional phrases and scenarios involving potential events or emergencies

No Comments

No Known Activations