INDEX

Explanations

words relating to personal experiences or perspectives

oai_token-act-pair · gpt-3.5-turbo

references to personal experiences and sentiments associated with positions of honor or privilege

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

link

-0.74

engine

-0.74

metadata

-0.74

olitics

-0.73

acists

-0.72

geist

-0.69

vae

-0.69

ohm

-0.66

Gi

-0.66

æ©Ł

-0.66

POSITIVE LOGITS

 witnessing

1.66

 owning

1.63

 navigating

1.59

 encountering

1.57

 watching

1.54

 seeing

1.51

 experiencing

1.49

 discovering

1.49

 being

1.48

 attending

1.45

Activations Density 0.478%

words relating to personal experiences or perspectives

references to personal experiences and sentiments associated with positions of honor or privilege

No Comments

No Known Activations