INDEX

Explanations

references to the concept of "prison"

oai_token-act-pair · gpt-3.5-turbo

references to prison

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

lass

-0.74

issan

-0.69

udden

-0.69

thora

-0.68

oric

-0.67

 Bundes

-0.66

laus

-0.66

yip

-0.65

///

-0.63

idy

-0.61

POSITIVE LOGITS

 prisons

0.93

prison

0.93

 inmates

0.92

 prison

0.92

 inmate

0.85

 barr

0.82

 confinement

0.82

 jail

0.81

 incarcer

0.80

 sentences

0.80

Activations Density 0.019%

references to the concept of "prison"

references to prison

No Comments

No Known Activations