INDEX

Explanations

the name "Durant," specifically with a higher activation for instances where the name is associated with positive or significant events

oai_token-act-pair · gpt-3.5-turbo

references to the basketball player Kevin Durant

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

ework

-0.78

LOAD

-0.74

tering

-0.72

bered

-0.72

scribed

-0.72

ocular

-0.70

glers

-0.70

actively

-0.68

course

-0.67

animate

-0.67

POSITIVE LOGITS

 Durant

0.88

 Westbrook

0.87

 Harden

0.85

minster

0.75

itect

0.70

ary

0.69

 Bros

0.69

KD

0.68

heimer

0.68

Dur

0.67

Activations Density 0.026%

the name "Durant," specifically with a higher activation for instances where the name is associated with positive or significant events

references to the basketball player Kevin Durant

No Comments

No Known Activations