INDEX

Explanations

references to specific information or sources mentioned in a text

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "references" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

aut

-0.69

orer

-0.67

letico

-0.64

anqu

-0.63

retch

-0.63

lot

-0.63

 ..........

-0.62

hap

-0.62

ieve

-0.62

ighty

-0.62

POSITIVE LOGITS

 references

3.78

 reference

2.49

 References

2.28

 referencing

1.85

 referenced

1.80

 mentions

1.69

reference

1.67

 quotations

1.66

 citations

1.64

Reference

1.58

Activations Density 0.020%

references to specific information or sources mentioned in a text

occurrences of the word "references" in various contexts

No Comments

No Known Activations