INDEX

Explanations

phrases that start with the word "a" or "an"

oai_token-act-pair · gpt-3.5-turbo

articles ("a" and "an") in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

weights

-0.69

antes

-0.69

osc

-0.68

asons

-0.67

TDs

-0.66

bugs

-0.66

enance

-0.66

izons

-0.66

flies

-0.65

ores

-0.65

POSITIVE LOGITS

 nutshell

1.37

 statement

0.94

 footnote

0.91

 twist

0.91

 flurry

0.88

 keynote

0.88

 perverse

0.84

 bizarre

0.82

 wake

0.80

 brief

0.79

Activations Density 0.063%

phrases that start with the word "a" or "an"

articles ("a" and "an") in the text

No Comments

No Known Activations