INDEX

Explanations

phrases indicating a current time frame or era

oai_token-act-pair · gpt-3.5-turbo

references to current trends or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

fect

-0.79

RTX

-0.77

 Hits

-0.76

regate

-0.72

onal

-0.71

henko

-0.70

edly

-0.69

agate

-0.68

idav

-0.65

ements

-0.65

POSITIVE LOGITS

adays

0.97

hift

0.84

days

0.82

pring

0.77

hops

0.75

abouts

0.71

paces

0.67

 everywhere

0.66

 transitioning

0.65

 undergoing

0.65

Activations Density 0.018%

phrases indicating a current time frame or era

references to current trends or conditions

No Comments

No Known Activations