INDEX

Explanations

examples of actions or events in various contexts

oai_token-act-pair · gpt-3.5-turbo

instances or cases described in a context that suggests variability or exceptionality

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

ãĤ´

-0.86

obin

-0.81

scribe

-0.81

ãĥ¥

-0.79

zar

-0.69

ãĤ´ãĥ³

-0.69

zag

-0.67

assi

-0.66

Äĩ

-0.65

agra

-0.64

POSITIVE LOGITS

 however

0.77

 including

0.76

 applicants

0.73

 moreover

0.71

 municip

0.65

 depending

0.64

 occurring

0.64

 claimants

0.64

 besides

0.63

 combinations

0.63

Activations Density 0.164%

examples of actions or events in various contexts

instances or cases described in a context that suggests variability or exceptionality

No Comments

No Known Activations