INDEX

Explanations

phrases related to the duration of events

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the duration of events or experiences

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

reci

-0.70

rim

-0.67

agon

-0.66

raised

-0.66

alty

-0.63

erning

-0.62

involved

-0.61

ledged

-0.60

 leveled

-0.58

ramid

-0.57

POSITIVE LOGITS

IER

0.82

 thro

0.77

 longer

0.77

 lasting

0.75

staking

0.74

Duration

0.73

stall

0.71

ime

0.69

inka

0.69

 satisf

0.68

Activations Density 0.015%

phrases related to the duration of events

phrases indicating the duration of events or experiences

No Comments

No Known Activations