INDEX

Explanations

phrases expressing hypothetical situations, assumptions, or possibilities

oai_token-act-pair · gpt-3.5-turbo

concepts related to impossibility and complex features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

atform

-0.71

olicy

-0.68

 Sonia

-0.65

ISION

-0.63

><

-0.62

rica

-0.61

rea

-0.59

iple

-0.58

istar

-0.57

cker

-0.57

POSITIVE LOGITS

 relate

0.77

 otherwise

0.76

 attributable

0.75

fit

0.75

fits

0.72

 elsewhere

0.71

 themselves

0.71

å¥

0.71

tain

0.70

 anat

0.69

Activations Density 0.550%

phrases expressing hypothetical situations, assumptions, or possibilities

concepts related to impossibility and complex features

No Comments

No Known Activations