INDEX

Explanations

references to public parks

oai_token-act-pair · gpt-3.5-turbo

mentions of parks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 dilig

-0.72

rontal

-0.70

xit

-0.69

 decomp

-0.68

ptoms

-0.67

nces

-0.64

initions

-0.63

theless

-0.62

soever

-0.62

 oxid

-0.62

POSITIVE LOGITS

park

1.02

 park

0.98

 ranger

0.85

hurst

0.84

wright

0.82

our

0.80

conservancy

0.80

keeper

0.79

way

0.77

keeping

0.76

Activations Density 0.014%

references to public parks

mentions of parks

No Comments

No Known Activations