INDEX

Explanations

terms related to physical approach or nearing

oai_token-act-pair · gpt-3.5-turbo

instances of the word "approaching" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

pite

-1.07

cell

-0.85

cake

-0.82

hal

-0.78

rd

-0.76

extra

-0.75

arus

-0.72

bel

-0.71

orie

-0.71

spons

-0.71

POSITIVE LOGITS

 Ceres

0.83

 approached

0.82

 Approach

0.71

 heights

0.71

 nearer

0.70

isons

0.69

 approach

0.69

IAS

0.67

WARD

0.67

heny

0.67

Activations Density 0.014%

terms related to physical approach or nearing

instances of the word "approaching" and its variations

No Comments

No Known Activations