INDEX

Explanations

words related to prevalence or frequency

oai_token-act-pair · gpt-3.5-turbo

instances of the word "common" and its variations in different contexts

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

mosp

-0.88

zik

-0.80

usalem

-0.75

etry

-0.70

=-=-

-0.70

forestation

-0.70

uana

-0.69

ophon

-0.67

oÄŁ

-0.67

ignt

-0.66

POSITIVE LOGITS

 occurrence

1.02

alities

1.01

 occurrences

1.01

places

0.98

ality

0.95

wealth

0.94

 denomin

0.90

place

0.77

pmwiki

0.74

Sym

0.72

Activations Density 0.029%

words related to prevalence or frequency

instances of the word "common" and its variations in different contexts

No Comments

No Known Activations