INDEX

Explanations

references to draining in a variety of contexts

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "drain" 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 @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

fman

-0.75

eering

-0.70

ordan

-0.69

HCR

-0.69

*/(

-0.65

ALE

-0.65

 Mortal

-0.64

PRES

-0.64

ourke

-0.64

eer

-0.61

POSITIVE LOGITS

 drain

0.96

pipe

0.85

teenth

0.83

pling

0.81

 draining

0.79

 basin

0.78

 drains

0.77

plings

0.74

 Drain

0.74

esville

0.72

Activations Density 0.030%

references to draining in a variety of contexts

references to the concept of "drain" in various contexts

No Comments

No Known Activations