INDEX

Explanations

special characters indicating a specific type of formatting or encoding in text

oai_token-act-pair · gpt-3.5-turbo

occurrences of the character "Ŀ"

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

 shape

-0.79

 warr

-0.79

 ende

-0.77

 undecided

-0.76

 likeness

-0.75

 imagination

-0.73

 unconscious

-0.73

 resemblance

-0.72

 unborn

-0.71

 barg

-0.70

POSITIVE LOGITS

ï¸ı

1.09

ï¸

1.00

Additionally

0.91

However

0.90

 Similarly

0.88

 Additionally

0.87

ttp

0.87

cue

0.86

Unless

0.84

Â¯

0.84

Activations Density 0.184%

special characters indicating a specific type of formatting or encoding in text

occurrences of the character "Ŀ"

No Comments

No Known Activations