INDEX

Explanations

terms related to "exiting" or leaving a situation

oai_token-act-pair · gpt-3.5-turbo

references to exit strategies or departures from a situation

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

elf

-0.82

acebook

-0.79

illac

-0.78

yright

-0.74

eta

-0.72

hed

-0.71

heet

-0.70

ened

-0.70

enegger

-0.70

sonian

-0.69

POSITIVE LOGITS

 spoiler

0.76

 exit

0.75

 pupil

0.72

 Door

0.72

 door

0.71

 Exit

0.67

 Spiegel

0.66

 doors

0.66

 Redemption

0.65

 velocity

0.63

Activations Density 0.015%

terms related to "exiting" or leaving a situation

references to exit strategies or departures from a situation

No Comments

No Known Activations