INDEX

Explanations

instances where time is mentioned or emphasized

oai_token-act-pair · gpt-3.5-turbo

duration-related phrases indicating the time it takes to complete various actions

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

alde

-0.95

ħĭ

-0.87

oeuv

-0.77

interstitial

-0.77

ntil

-0.74

inged

-0.73

ailability

-0.72

disabled

-0.70

ixture

-0.70

idden

-0.70

POSITIVE LOGITS

 approx

0.74

hrs

0.69

 hectares

0.68

 Hess

0.68

 msec

0.68

 wrench

0.67

 Sims

0.67

 Horowitz

0.67

(~

0.66

 Lyft

0.64

Activations Density 0.126%

instances where time is mentioned or emphasized

duration-related phrases indicating the time it takes to complete various actions

No Comments

No Known Activations