INDEX

Explanations

time-related phrases with a focus on the remaining time for a task or event

oai_token-act-pair · gpt-3.5-turbo

phrases indicating time constraints or remaining duration

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

tnc

-0.94

cult

-0.73

advertising

-0.71

isms

-0.70

usa

-0.70

effects

-0.69

doc

-0.69

ãĤ´ãĥ³

-0.69

SourceFile

-0.68

çĶ°

-0.68

POSITIVE LOGITS

 apiece

1.09

 allotted

0.76

 elapsed

0.74

veyard

0.73

per

0.70

 total

0.69

 each

0.67

 devoted

0.65

onies

0.65

 dedicated

0.63

Activations Density 0.227%

time-related phrases with a focus on the remaining time for a task or event

phrases indicating time constraints or remaining duration

No Comments

No Known Activations