INDEX

Explanations

phrases related to things that are not unexpected

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a lack of surprise or expectation

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

abase

-0.91

ngth

-0.86

 raft

-0.77

xe

-0.75

sembly

-0.72

ailability

-0.72

 negoti

-0.71

ictionary

-0.71

alach

-0.71

 bowling

-0.69

POSITIVE LOGITS

DragonMagazine

0.76

imaru

0.71

Pav

0.67

 anymore

0.66

 Mellon

0.65

actionDate

0.65

Notice

0.63

 systemd

0.63

 considering

0.63

à¤¾

0.63

Activations Density 0.040%

phrases related to things that are not unexpected

phrases indicating a lack of surprise or expectation

No Comments

No Known Activations