INDEX

Explanations

phrases related to time and events happening in sequence

oai_token-act-pair · gpt-3.5-turbo

references to events or situations that occur after a significant action or moment

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

inho

-0.74

ashtra

-0.73

quote

-0.73

QL

-0.70

IOR

-0.68

IUM

-0.67

LESS

-0.66

<?

-0.65

ãĥĺãĥ©

-0.64

ELY

-0.63

POSITIVE LOGITS

 arrival

0.81

 initial

0.77

 completion

0.72

 revelation

0.72

noon

0.71

 exhausting

0.71

 unsuccessful

0.71

 stal

0.69

 expiration

0.68

 incident

0.68

Activations Density 0.251%

phrases related to time and events happening in sequence

references to events or situations that occur after a significant action or moment

No Comments

No Known Activations