INDEX

Explanations

words related to the past tense of actions or events

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the occurrence and timing of events

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

hirt

-0.78

 endors

-0.75

 endorsements

-0.72

orah

-0.70

 outfit

-0.69

 wearer

-0.69

otton

-0.69

ded

-0.68

iculty

-0.67

beard

-0.67

POSITIVE LOGITS

ï¸

0.85

 Wonderland

0.75

Interstitial

0.75

 millisec

0.72

 limbo

0.71

GMT

0.71

UTC

0.70

 Yugoslavia

0.69

 interstate

0.69

 Phase

0.69

Activations Density 0.261%

words related to the past tense of actions or events

phrases indicating the occurrence and timing of events

No Comments

No Known Activations