INDEX

Explanations

references to how individuals are dressed

oai_token-act-pair · gpt-3.5-turbo

references to people being dressed 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 @ 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

venants

-0.78

ellen

-0.70

found

-0.69

pelling

-0.66

 depress

-0.65

orb

-0.65

SHIP

-0.64

rw

-0.64

heric

-0.63

prus

-0.63

POSITIVE LOGITS

glers

0.87

 rehearsal

0.84

uce

0.83

 gown

0.81

 attire

0.81

 rehears

0.78

 shirts

0.77

 dresses

0.76

 dressed

0.75

 uniforms

0.75

Activations Density 0.019%

references to how individuals are dressed

references to people being dressed in various contexts

No Comments

No Known Activations