INDEX

Explanations

personal pronouns followed by verbs indicating action

oai_token-act-pair · gpt-3.5-turbo

pronouns and personal references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 srfAttach

-0.82

Elsewhere

-0.79

 Others

-0.79

Others

-0.75

 Amen

-0.75

 Else

-0.73

soType

-0.72

Another

-0.71

 Additional

-0.71

 Lastly

-0.69

POSITIVE LOGITS

 literally

0.91

've

0.82

asma

0.80

 WANT

0.78

 invented

0.76

 want

0.76

 basically

0.75

 simply

0.73

 genuinely

0.73

 relentlessly

0.72

Activations Density 0.409%

personal pronouns followed by verbs indicating action

pronouns and personal references

No Comments

No Known Activations