INDEX

Explanations

phrases suggesting options or alternatives

oai_token-act-pair · gpt-3.5-turbo

phrases that suggest alternatives or conditional scenarios

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

IDENT

-0.69

emi

-0.68

owner

-0.67

estro

-0.64

onday

-0.64

ETS

-0.62

igr

-0.61

ires

-0.61

UD

-0.61

ursday

-0.60

POSITIVE LOGITS

chard

1.23

 alternatively

1.17

Else

1.05

 else

0.98

acular

0.94

nery

0.93

 whatever

0.88

 otherwise

0.88

 maybe

0.85

 worse

0.85

Activations Density 0.065%

phrases suggesting options or alternatives

phrases that suggest alternatives or conditional scenarios

No Comments

No Known Activations