INDEX

Explanations

phrases related to reading or textual content

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Reading," indicating a focus on discussions about reading materials or activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ortality

-0.90

xon

-0.83

ella

-0.74

ega

-0.72

Ø§ÙĦ

-0.71

pload

-0.69

ascal

-0.69

orp

-0.69

WWF

-0.67

VC

-0.66

POSITIVE LOGITS

 comprehension

1.23

 aloud

1.13

just

0.89

ied

0.87

 reading

0.79

 glasses

0.78

 readings

0.77

ying

0.76

 Reading

0.73

mitt

0.73

Activations Density 0.022%

phrases related to reading or textual content

instances of the word "Reading," indicating a focus on discussions about reading materials or activities

No Comments

No Known Activations