INDEX

Explanations

phrases related to events or announcements

oai_token-act-pair · gpt-3.5-turbo

the article "a" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ATURES

-0.79

docs

-0.67

items

-0.66

rils

-0.64

adding

-0.63

baugh

-0.62

coins

-0.61

alties

-0.61

 talents

-0.61

Ã©r

-0.61

POSITIVE LOGITS

 glance

1.03

 concess

0.90

 behest

0.89

 disadvantage

0.87

 leisure

0.85

 expense

0.85

 junction

0.84

 whopping

0.82

 checkpoint

0.82

 whim

0.81

Activations Density 0.072%

phrases related to events or announcements

the article "a" in various contexts

No Comments

No Known Activations