INDEX

Explanations

phrases indicating time or sequence of events

oai_token-act-pair · gpt-3.5-turbo

sequences that indicate a passage of time or sequential events

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

SON

-0.76

ãĤ´ãĥ³

-0.73

gow

-0.71

quartered

-0.69

own

-0.67

erver

-0.66

zar

-0.64

role

-0.63

atically

-0.62

arch

-0.62

POSITIVE LOGITS

 however

0.90

 though

0.71

 meanwhile

0.71

 when

0.68

emetery

0.67

 Hiroshima

0.65

 Basil

0.65

 Collider

0.63

 alas

0.61

 Verd

0.59

Activations Density 0.359%

phrases indicating time or sequence of events

sequences that indicate a passage of time or sequential events

No Comments

No Known Activations