INDEX

Explanations

phrases related to activities or events happening for a specified duration of time

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

Rules

-0.82

marks

-0.80

upon

-0.72

Cho

-0.69

Enjoy

-0.69

Att

-0.68

anism

-0.68

flows

-0.68

via

-0.67

 Originally

-0.67

POSITIVE LOGITS

 sake

1.26

 multitude

0.99

 foreseeable

0.97

 bunch

0.92

 glimpse

0.91

 hypothetical

0.89

 variety

0.89

 plethora

0.88

new

0.87

 whopping

0.86

Activations Density 0.140%

phrases related to activities or events happening for a specified duration of time

occurrences of the article "a."

No Comments

No Known Activations