INDEX

Explanations

positive comments or praise related to strength or performance

oai_token-act-pair · gpt-3.5-turbo

expressions of appreciation or admiration

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 blamed

-0.81

CLAIM

-0.78

 inconvenient

-0.77

 fearing

-0.77

 lest

-0.76

claimer

-0.76

unless

-0.75

Myth

-0.74

soDeliveryDate

-0.74

ģĸ

-0.73

POSITIVE LOGITS

 teamwork

1.06

 professionalism

1.03

 ingenuity

0.93

 caliber

0.93

 versatility

0.92

 strides

0.92

 rapport

0.90

 synergy

0.88

 accomplishments

0.87

 turnout

0.87

Activations Density 0.428%

positive comments or praise related to strength or performance

expressions of appreciation or admiration

No Comments

No Known Activations