INDEX

Explanations

phrases indicating the attainment of benefits, rewards, or positive outcomes

oai_token-act-pair · gpt-3.5-turbo

occurrences of the article "a"

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

anism

-0.92

 endeavors

-0.74

Edit

-0.74

CI

-0.73

 Measures

-0.71

ie

-0.69

agree

-0.68

aneously

-0.68

Area

-0.68

livious

-0.66

POSITIVE LOGITS

lot

1.36

 bunch

1.12

 glimpse

1.07

 handful

1.02

 plethora

1.02

few

0.98

 huge

0.98

 couple

0.92

 sizeable

0.92

 whopping

0.92

Activations Density 0.330%

phrases indicating the attainment of benefits, rewards, or positive outcomes

occurrences of the article "a"

No Comments

No Known Activations