INDEX

Explanations

the word "glimpse" followed by a preposition or "into"

oai_token-act-pair · gpt-3.5-turbo

phrases related to getting a view or insight into something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

raid

-0.91

die

-0.83

iction

-0.71

currency

-0.70

death

-0.70

iled

-0.70

fighting

-0.69

erto

-0.67

queue

-0.65

ded

-0.65

POSITIVE LOGITS

 glimps

1.12

seeing

0.98

 glimpse

0.95

 hints

0.85

 peek

0.81

ewitness

0.75

 hint

0.74

terday

0.73

finder

0.73

 clues

0.70

Activations Density 0.027%

the word "glimpse" followed by a preposition or "into"

phrases related to getting a view or insight into something

No Comments

No Known Activations