INDEX

Explanations

times or durations indicated by the word 'During'

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a time frame or period of events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

govtrack

-0.79

ould

-0.79

raid

-0.72

optional

-0.69

tools

-0.69

lor

-0.68

rail

-0.67

adr

-0.66

onian

-0.65

igans

-0.64

POSITIVE LOGITS

itialized

0.86

 newsp

0.86

 veter

0.78

ventory

0.76

ORPG

0.72

During

0.72

 During

0.70

quickShipAvailable

0.70

 periods

0.69

 evaluating

0.68

Activations Density 0.009%

times or durations indicated by the word 'During'

phrases indicating a time frame or period of events

No Comments

No Known Activations