INDEX

Explanations

words related to cause and effect, specifically consequences

oai_token-act-pair · gpt-3.5-turbo

phrases related to consequences of actions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

nas

-0.73

bers

-0.67

bull

-0.65

staff

-0.60

bus

-0.59

brew

-0.59

berman

-0.58

ba

-0.57

Base

-0.57

Nut

-0.57

POSITIVE LOGITS

 consequence

3.79

 reperc

1.92

 result

1.92

 consequences

1.84

 implication

1.64

 consequ

1.43

 repercussions

1.41

 outcome

1.34

 ramifications

1.31

 complication

1.30

Activations Density 0.006%

words related to cause and effect, specifically consequences

phrases related to consequences of actions or events

No Comments

No Known Activations