INDEX

Explanations

phrases related to unexpected appearances or arrivals

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "show up."

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

cest

-0.86

aga

-0.65

coh

-0.65

quez

-0.61

bis

-0.61

imar

-0.60

ided

-0.60

sen

-0.59

imov

-0.59

imb

-0.58

POSITIVE LOGITS

stairs

0.75

ngth

0.74

doors

0.71

oon

0.69

 unexpectedly

0.68

 here

0.66

 hither

0.66

 prominently

0.64

 smelling

0.64

number

0.63

Activations Density 0.035%

phrases related to unexpected appearances or arrivals

instances of the phrase "show up."

No Comments

No Known Activations