INDEX

Explanations

phrases containing the word "predictable."

oai_token-act-pair · gpt-3.5-turbo

terminology related to predictability and unpredictability

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

arte

-0.85

burgh

-0.83

gur

-0.81

tha

-0.79

agine

-0.77

aina

-0.73

sten

-0.72

ovember

-0.71

onia

-0.71

stones

-0.70

POSITIVE LOGITS

 behavior

0.81

 behaviour

0.79

 Predict

0.74

aneously

0.73

 unpredict

0.72

 intervals

0.72

 outcomes

0.71

ãĥĻ

0.70

 punishments

0.70

ively

0.69

Activations Density 0.022%

phrases containing the word "predictable."

terminology related to predictability and unpredictability

No Comments

No Known Activations