INDEX

Explanations

phrases indicating a cause and effect relationship

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate causes and effects or the act of bringing about changes

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

schild

-0.82

 Seym

-0.71

dating

-0.71

livious

-0.69

raid

-0.66

codes

-0.66

rics

-0.63

debian

-0.61

rant

-0.61

doms

-0.60

POSITIVE LOGITS

 forth

0.99

 forward

0.79

endum

0.78

 together

0.75

bring

0.74

up

0.67

udic

0.66

hurst

0.65

upp

0.65

ageddon

0.65

Activations Density 0.033%

phrases indicating a cause and effect relationship

phrases that indicate causes and effects or the act of bringing about changes

No Comments

No Known Activations