INDEX

Explanations

phrases related to links or sources for more information

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "for" 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 @ 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

orbit

-0.85

nil

-0.73

intention

-0.72

adr

-0.69

iltr

-0.69

é¾

-0.67

church

-0.67

illin

-0.66

oln

-0.66

ensibly

-0.65

POSITIVE LOGITS

 example

1.01

 instance

0.98

bidden

0.93

 starters

0.86

 updates

0.80

 FREE

0.79

 reasons

0.79

geries

0.77

 clarification

0.77

gery

0.75

Activations Density 0.089%

phrases related to links or sources for more information

occurrences of the word "for" in various contexts

No Comments

No Known Activations